[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_125":126},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module not found","tools","module",{"_20":22,"_23":21,"_24":25,"_26":27,"_28":29,"_30":31,"_32":33,"_34":25,"_35":36,"_37":-5,"_38":-5,"_39":-5,"_40":41,"_42":43,"_44":45,"_46":47,"_48":49,"_50":51,"_52":53,"_54":-5,"_55":47,"_56":57,"_58":59,"_60":-5,"_61":-5,"_62":-5,"_63":-5,"_64":65,"_66":67,"_68":69,"_70":71,"_72":73,"_74":-5,"_75":-5,"_76":-5,"_77":47,"_78":-5,"_79":80,"_81":82,"_83":84,"_85":86,"_87":88,"_89":90,"_91":92},"apiUrl","https://odel.app/api","creator","tenequm","moduleId","pond","@tenequm/pond","displayName","tagline","Lossless archive and search for AI agent sessions across clients, exposed to agents over MCP.","iconKey","https://avatars.githubusercontent.com/u/22403766?v=4","category","local","status","published","visibility","public","metaDescription","deploymentType","external","workerName","bindingName","mcpConfig","pricingModel","free","creatorUserId","system:external-import","createdAt",1786171375000,"updatedAt",1788167611000,"sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=io.github.tenequm%2Fpond","sourceVerified",true,"sourceMissingSince","importedAt","upstreamStars",57,"upstreamUpdatedAt",1788166433000,"detectedTransport","detectedAuthType","probeStatus","lastProbedAt","githubRepoUrl","https://github.com/tenequm/pond","githubRepoFullName","tenequm/pond","releaseId","ext-tenequm-pond","version","0.14.5","descriptionMarkdown","

\n $\"pond\"$ \n

\n\n# pond\n\n[![CI](https://img.shields.io/github/actions/workflow/status/tenequm/pond/ci.yml?branch=main&style=flat-square)](https://github.com/tenequm/pond/actions/workflows/ci.yml)\n[![crates.io](https://img.shields.io/crates/v/pond-db.svg?style=flat-square)](https://crates.io/crates/pond-db)\n[![docs](https://img.shields.io/badge/docs-pond.locker-blue?style=flat-square)](https://pond.locker/)\n[![License](https://img.shields.io/badge/license-Apache--2.0-blue.svg?style=flat-square)](LICENSE)\n\n> \"I know we discussed that before. Why can't I find that damn conversation?\"\n\nPond makes every AI agent session you've ever run - Claude Code, Codex, any tool, any machine - searchable in one place.\n\nYour agent history is already on your disk: thousands of sessions full of decisions, fixes, and dead ends - scattered across tools that can't search them. Pond ingests them all automatically and losslessly into storage you own (a local dir or your own S3 bucket), makes the whole corpus searchable and SQL-queryable, and hands that recall back to your agents over MCP - so \"how did we fix this before?\" is a query, not an archaeology dig. Sessions stop being locked to the tool that created them: any session can be restored into any supported client and continued there.\n\n```sh\nbrew install tenequm/tap/pond # macOS / Linux\n\nscoop bucket add tenequm https://github.com/tenequm/scoop-bucket # Windows\nscoop install tenequm/pond\n```\n\nOr prompt your agent: *\"Please install and set up pond (see github.com/tenequm/pond)\"* - the full, failure-proofed version of that prompt is in [Connect your agents](https://pond.locker/get-started/connect-your-agents#agent-setup-prompt).\n\n

\n $\"A$ \n

_{A live 12k-session corpus, then a three-month-old fix found and verified against the current code (crisper MP4)}

\n\n

\n \n \n $\"Bar$ \n \n

_{Five real recall questions, one corpus, one machine - method, raw numbers, and a rerunnable script in docs/benchmarks}

\n\nStatus: pre-v1. Schemas, wire shapes, and config keys are subject to breaking change until v1. Full documentation lives at [pond.locker](https://pond.locker/); the contract is [`docs/spec.md`](https://github.com/tenequm/pond/blob/main/docs/spec.md).\n\n## Quickstart\n\nInstall, run guided setup, and ingest your local sessions:\n\n```sh\nbrew install tenequm/tap/pond # macOS / Linux; Windows: Scoop or a release zip - see Install below\npond init # guided setup: storage, adapters, MCP + agent skill, optional schedule\npond sync # ingest and index - every enabled adapter\n```\n\n`pond init` registers pond as an MCP server for Claude Code and installs the bundled pond skill; for Codex it prints the command to run instead - restart the client afterwards so the tools load. By hand: `claude mcp add -s user pond -- pond mcp`, `codex mcp add pond -- pond mcp`; skill: `mkdir -p ~/.claude/skills/pond && pond skill > ~/.claude/skills/pond/SKILL.md` (that whole line is POSIX-only - `mkdir -p`, `&&`, and `>` all break or corrupt in Windows PowerShell 5.1; use the PowerShell block in [Connect your agents](https://pond.locker/get-started/connect-your-agents)). Then ask your agent - real prompts from daily use:\n\n```\ncheck in pond how we solved this before, then apply the same fix here\n```\n```\nwhere we left off yesterday - check pond, then continue\n```\n```\nare you sure that won't break X? check in pond how we struggled with exactly this\n```\n\nSessions are picked up automatically from **Claude Code**, the **Claude desktop app** (local agent mode), **Codex CLI**, **opencode**, **pi-coding-agent**, **oh-my-pi**, **OpenClaw**, **NanoClaw**, **Hermes Agent**, **letta-code**, and **grok-build**. A Claude.ai data export imports with `pond sync claude-ai-export --path ` (manual download, so not auto-discovered).\n\n## Isn't this another memory tool?\n\nNo - it's the layer underneath one. Memory tools store what they decided you'd need - facts, summaries, filed chunks; the sessions themselves are gone. Pond keeps the sessions: every message, tool call, and result, value-complete, cross-client, in storage you own, never pruned - searchable over MCP and restorable into any client. Memory is a derived view you can always rebuild from an archive; an archive can never be rebuilt from memories.\n\nThree kinds of tool get called \"memory\". Side by side:\n\n| | **pond** | **Session search** (ctx, deja-vu, cass) | **Memory layers** (Mem0, Letta) |\n|---|---|---|---|\n| History from before install | yes | yes | no |\n| What is kept | the whole session | a search index ¹ | extracted facts |\n| Where it lives | local dir or S3 bucket | local index | the tool's database |\n| After the harness deletes the file | still there | gone at next refresh ¹ | only the extract |\n| Several machines | one shared bucket | pulled into one machine ² | shared server or cloud |\n| Agent access | CLI, MCP, HTTP, SQL | CLI, MCP ³ | HTTP, SDK, MCP |\n\n¹ cass also mirrors the raw files, so they outlive the source. ² deja-vu copies records between machines over ssh; cass pulls other hosts' session files over ssh/rsync into its local index; ctx is single-machine. ³ cass has no MCP server.\n\nPick session search for fast local recall. Pick a memory layer when the agent should carry distilled facts, not the record. Pick pond when you want the sessions themselves, in storage you own, from every machine you run.\n\nFull comparison, with receipts per tool: [pond.locker/compare](https://pond.locker/compare). Every cell is a claim about a specific version of someone else's project. If one has gone stale, [open an issue](https://github.com/tenequm/pond/issues) and it gets fixed the same day.\n\n## Background\n\nEvery agentic CLI ships its own session format and its own search surface. Switching tools means losing history. Replaying a Claude Code session in another provider's tooling means re-translating the wire shape by hand. Hosted multi-tenant deployments rebuild the same storage layer from scratch.\n\nPond is the storage and retrieval layer that sits underneath. Every adapter is a bidirectional codec between a client format and one canonical schema, so any session can be restored by any adapter - it need not return to the client that produced it. Storage, search (BM25 full-text by default, with optional semantic search, one arm per query), and provider-agnostic replay all sit on a single Lance-on-object-storage foundation.\n\nThe v1 surface includes: full CLI, HTTP+JSON and MCP transports, search over three Lance datasets, opt-in `intfloat/multilingual-e5-small` embeddings at FP16 weights (Metal on macOS, CUDA opt-in, CPU fallback), and local-FS / S3 / GCS / Azure backends through Lance's `object_store` integration.\n\n## Install\n\nLinux, macOS, and Windows are supported.\n\n**macOS and Linux:**\n\n```sh\nbrew install tenequm/tap/pond # Homebrew\nnix profile add github:tenequm/pond#pond # Nix\n```\n\n**Windows** (Scoop, the primary channel - it also ships `pondw.exe`, the windowless launcher that scheduled sync runs through):\n\n```powershell\nscoop bucket add tenequm https://github.com/tenequm/scoop-bucket\nscoop install tenequm/pond\n```\n\nBuckets are git clones, so the first line needs git on `PATH` - if it fails with \"Git is required for buckets\", run `scoop install git` and retry.\n\nNo Scoop yet? Bootstrap it first, from a **normal (non-admin) PowerShell** (its installer refuses an elevated shell):\n\n```powershell\nSet-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser -Force\nirm get.scoop.sh | iex\n```\n\nThen open a new terminal - `PATH` changes reach only processes started after the install. See the [Windows notes](https://pond.locker/get-started/install#windows) for Defender, long paths, scheduling, and WSL.\n\n**No package manager (any platform):** every [release](https://github.com/tenequm/pond/releases) attaches prebuilt binaries (`pond-x86_64-pc-windows-msvc.zip` on Windows, ~223 MB unpacked) - unpack one and add its directory to `PATH` (on Windows: Settings > System > About > Advanced system settings > Environment Variables, under your user variables), nothing else to install.\n\n**Via cargo (any platform, needs the Rust toolchain):**\n\n```sh\ncargo binstall pond-db # downloads the prebuilt binary (needs cargo-binstall)\ncargo install pond-db # builds from crates.io (installs the `pond` command)\n```\n\nBoth install `pond.exe` only, so scheduled sync on Windows - which runs through `pondw.exe` - wants the Scoop or zip install instead (or `--features windows-launcher` on a source build). `cargo install` also needs the protoc and NASM prerequisites below.\n\n**Build from source:**\n\n```sh\ngit clone https://github.com/tenequm/pond.git\ncd pond\ncargo install --path packages/pond\n```\n\nOn Windows that clone needs `git config --global core.longpaths true` first (test-fixture paths exceed 260 characters), and the build needs an explicit `--target x86_64-pc-windows-msvc` - without it cargo applies the repo's `+crt-static` flag to build scripts and proc-macros too, which then fail to load.\n\nFor CUDA acceleration on Linux:\n\n```sh\ncargo install --path packages/pond --features cuda\n```\n\nOn macOS the Metal backend is selected automatically; on other systems the CPU fallback runs without extra features. Building from source on Windows additionally needs `protoc` (`winget install Google.Protobuf`) and NASM (`winget install NASM.NASM`, then add `C:\\Program Files\\NASM` to `PATH` yourself - its installer doesn't) on `PATH`.\n\nIf any install path fails, the [Troubleshooting guide](https://pond.locker/guides/troubleshooting) covers the common stalls.\n\n## Usage\n\n### Sync and search\n\nSet up storage, adapters, MCP registration, and an optional sync schedule in one pass (idempotent - re-run it any time to repair or update):\n\n```sh\npond init\n```\n\nThen import sessions from local adapters, update indexes, and search:\n\n```sh\npond sync\npond search \"how did we wire up the OCC retry loop\"\n```\n\n### Run a server\n\n```sh\npond serve # HTTP on 127.0.0.1:9797\npond serve --transport stdio # MCP over stdio\npond mcp # alias for stdio MCP\n```\n\n### Fetch and copy\n\nFetch a single session or message, or move a whole corpus:\n\n```sh\npond get-session \npond get-message \npond copy --from local --to snapshot.pond\npond copy --from snapshot.pond --to local\n```\n\n### Read-only SQL\n\nAsk structured questions with read-only SQL (the same surface as the `pond_sql` MCP tool):\n\n```sh\npond sql \"SELECT project, count(*) FROM messages GROUP BY project ORDER BY 2 DESC\"\n```\n\n### Maintenance\n\nRun maintenance on demand (sync folds indexes on every run):\n\n```sh\npond optimize --only embed # only when [embeddings].enabled = true\npond optimize --only index\n```\n\n### Scheduled sync\n\nKeep pond current automatically (launchd on macOS, systemd user timers or cron on Linux, Task Scheduler on Windows - there, run it from a normal shell, not an elevated one, or the task ends up owned by Administrators):\n\n```sh\npond schedule start # every 5m by default (--every 15m|1h|6h|1d)\npond schedule status\npond schedule logs\n```\n\n### Status and introspection\n\n`pond status` prints a per-table storage table, then `indexes` (text readiness, plus the semantic half only when embeddings are enabled), `stored` (sessions + messages), `agents` (source agents in the store), and this host's view of it: per-adapter sessions pending sync, the last sync's outcome (including a surfaced failure from a scheduled run), and the next scheduled run. `pond status --hosts` breaks a shared store down by ingest host; `--include-subagents` counts each subagent as its own agent. `pond sync --dry-run` previews what the next sync would read. `pond search --explain` returns Lance's `analyze_plan` output for each retrieval arm.\n\n### Remote storage\n\nBy default pond stores data locally under `~/.local/share/pond` (`%LOCALAPPDATA%\\pond\\data` on Windows). To use an object store, add credentials and switch the destination:\n\n```sh\npond creds add # interactive: name, access key, hidden secret\npond storage use s3+https://nbg1.your-objectstorage.com/my-pond # probe end-to-end, then flip [storage].path\npond storage check # verify: parse, creds, conditional-put (OCC), write/read/delete\n```\n\n`pond init --storage-path ` configures a remote destination during setup and prompts for credentials inline when the destination is remote, so a bucket is one command. The `s3+https://host/bucket` form works for any S3-compatible store (Hetzner, R2, B2, MinIO); `s3://`, `gs://`, and `az://` use the standard cloud SDK credential chain when no `[creds.*]` set matches. `pond copy --from --to ` carries existing local data into the bucket - idempotent, never deletes the source, and on completion it rebuilds the destination indexes and verifies every row landed (exit 6 if any are missing or duplicated, so you never reconcile by hand). `pond copy --verify-only --from --to ` runs that same check read-only, without copying. Full walkthrough: [pond.locker](https://pond.locker/).\n\n### Configuration\n\n`pond init` walks through everything below interactively and enables the adapters it finds. `pond sync` only ingests already-enabled adapters - enabling one is an explicit step (`pond adapters enable` / `pond adapters discover` / `pond init`), never a side effect of sync. Config lives at `~/.config/pond/config.toml` on macOS and Linux, `%APPDATA%\\pond\\config.toml` on Windows (`pond config path` prints it). Every `[adapters.]` block needs `enabled = true` to be active; sections without it (or with `enabled = false`) are skipped. `~` in paths expands on every platform (`%USERPROFILE%` on Windows).\n\n```toml\n[adapters.claude-code]\nenabled = true\npath = \"~/.claude/projects\"\n\n[adapters.codex-cli]\nenabled = false # kept in config, skipped on `pond sync`\npath = \"~/.codex/sessions\"\n```\n\nSearch is BM25 full-text by default. Semantic search is opt-in and off unless you ask for it: with it off no pond process downloads or loads an embedding model, new messages get no vectors, and `--mode vector` is refused. Turn it on in config or with `POND_EMBEDDINGS_ENABLED=true`, then run `pond optimize --only embed` once to fill the backlog:\n\n```toml\n[embeddings]\nenabled = true\n```\n\nFull detail, including what it costs and how mixed fleets behave, is in the [configuration reference](https://pond.locker/reference/configuration).\n\n### Supported harnesses\n\nOne adapter per harness, in `pond adapters` discovery order; `Reads` is the path `pond init` discovers and writes to `[adapters.].path` - shown POSIX-style, with the same home-relative layout on Windows (`~/.claude/projects` is `%USERPROFILE%\\.claude\\projects`). `Last verified` is the most recent capture or refresh date of the adapter's committed fixture (`packages/pond/tests/fixtures/adapter/`), the corpus its mapping is tested against. Adapters are maintained best-effort, and format drift is safe by design: unknown record shapes still ingest losslessly, malformed input surfaces as a typed error naming the file. Adding a harness is routine work - see [Contributing](#contributing).\n\n| Adapter | Reads | Last verified |\n|---------|-------|---------------|\n| `claude-code` | Claude Code CLI, `~/.claude/projects` | 2026-08-14 |\n| `claude-desktop-app` | Claude Desktop / Cowork local agent sessions | 2026-05-13 |\n| `claude-ai-export` | claude.ai data-export archive (manual `--path`) | 2026-06-04 |\n| `codex-cli` | OpenAI Codex CLI, `~/.codex/sessions` | 2026-05-13 |\n| `opencode` | opencode, `~/.local/share/opencode` (SQLite DB + legacy tree) | 2026-07-14 |\n| `openclaw` | openclaw, `~/.openclaw` | 2026-05-13 |\n| `nanoclaw` | nanoclaw, `~/nanoclaw` (the install root holding `data/v2-sessions`) | 2026-05-14 |\n| `hermes` | Hermes Agent, `~/.hermes` (`state.db` per profile) | 2026-07-23 |\n| `pi-coding-agent` | pi, `~/.pi/agent/sessions` | 2026-08-06 |\n| `oh-my-pi` | oh-my-pi, `~/.omp/agent/sessions` (ingest-only) | 2026-08-14 |\n| `letta-code` | letta-code, `~/.letta/transcripts` (a root relocated via `LETTA_TRANSCRIPT_ROOT` is configured as an explicit `path`) | 2026-08-24 |\n| `grok-build` | grok-build (xAI `grok` CLI), `~/.grok/sessions` (a relocated `GROK_HOME` is configured as an explicit `path`) | 2026-08-24 |\n\n### Verbosity\n\nRoot-level `-v` / `-vv` / `-vvv` raise the tracing level (info / debug / trace); `-q` / `-qq` lower it. The default surfaces warnings only. `RUST_LOG` overrides the CLI flag when set; `POND_LOG` is no longer honored.\n\n## Design\n\nThe full contract is in [`docs/spec.md`](https://github.com/tenequm/pond/blob/main/docs/spec.md). Key choices:\n\n- **Lance direct, no wrapper.** The `lance-format/lance` crates are the only storage and search engine. No `lancedb`, no parallel abstraction. Storage, indexing, OCC, schema evolution, blob columns, versioning, and time-travel are all Lance. The read-only `pond sql` surface is DataFusion planning over the same Lance datasets - a query escape hatch, not a second engine.\n- **Canonical Session / Message / Part interlingua.** Owned in pond, in the shape of Effect v4's `Prompt`-side Part union. This schema is pond's product; everything else is machinery around it.\n- **Three Lance datasets** (`sessions`, `messages`, `parts`). `messages` carries the nullable embedding (`vector` + `embedding_model`) alongside denormalized filter columns (`source_agent` / `project` / `role` / `timestamp`) for single-stage filter pushdown.\n- **No-synthesis adapter seam.** Adapters parse source records through extractor helpers that make \"invent a value\" a compile error - `model-no-synthesis`, `model-schema-honesty`, and `adapter-provenance-required` are structural, not review rules.\n- **Index lifecycle decoupled from writes.** Writes commit data (including embeddings, computed inline at ingest when embeddings are enabled) without folding the search indexes. `pond sync` runs index maintenance by default, and `pond optimize --only index` runs it on demand; Lance merges index results with a flat scan over unindexed fragments, so reads stay correct.\n- **Single-arm retrieval.** Each query runs one retriever - `fts` (BM25, the default) or `vector` (cosine, with a gentle recency tiebreaker, offered only when embeddings are enabled) - chosen per query; no server-side fusion. `--sort-by recency` returns newest-first. Results group to one summary per session, keyed on `session_root`.\n- **Language-neutral full-text.** Word-level `simple` tokenizer with English stemming (ascii-folding on); tokens the stemmer does not recognize pass through unchanged and stay exact-matchable, so pond indexes sessions in any language alike.\n- **Two transports, one handler set.** HTTP+JSON (axum) and MCP (rmcp) both dispatch into the same handlers. Wire ops: `pond_search`, `pond_get_session`, `pond_get_message`, `pond_ingest`. MCP additionally exposes the read-only `pond_sql` tool and the `schema://pond`, `schema://pond-sql`, and `stats://pond` resources.\n- **Opaque-string multi-tenancy.** Each tenant is a `namespace` string the integrator supplies; pond does not authenticate, authorize, or model identity. The object store's IAM is the storage boundary.\n- **Encryption is operational.** Bucket SSE plus filesystem encryption; pond holds no keys and adds no application-level crypto.\n\n## Roadmap\n\npond ships in small steps. This table lists the steps in order. Done steps stay in the table. The [roadmap board](https://github.com/users/tenequm/projects/1) holds the same items with their issues. React or comment on an issue to influence the order.\n\n| # | Step | Status |\n|---|------|--------|\n| 1 | Lossless ingest from Claude Code and Codex into Lance, local or S3 | ✅ [v0.5](https://github.com/tenequm/pond/releases/tag/v0.5.0) |\n| 2 | Single-arm search: BM25 or vector, one arm per query | ✅ [v0.10](https://github.com/tenequm/pond/releases/tag/v0.10.0) |\n| 3 | Remote sync in under a minute; warm search in under a second | ✅ [v0.11](https://github.com/tenequm/pond/releases/tag/v0.11.0)-[v0.12](https://github.com/tenequm/pond/releases/tag/v0.12.0) |\n| 4 | Tool-call columns and read-only SQL over the corpus | ✅ [v0.13](https://github.com/tenequm/pond/releases/tag/v0.13.0) |\n| 5 | Crash-safe local stores that self-heal on open | ✅ [v0.14.0](https://github.com/tenequm/pond/releases/tag/v0.14.0) |\n| 6 | Eleven harnesses, `pond resume` into any client, MCP registry, Windows | ✅ [v0.14.11](https://github.com/tenequm/pond/releases/tag/v0.14.11) |\n| 7 | BM25 becomes the default arm. Embeddings become opt-in. [#164](https://github.com/tenequm/pond/issues/164) | ✅ [v0.15.0](https://github.com/tenequm/pond/releases/tag/v0.15.0) |\n| 8 | New adapters become routine: `add-adapter` playbook + conformance harness [#172](https://github.com/tenequm/pond/issues/172), letta-code [#170](https://github.com/tenequm/pond/issues/170), grok CLI [#171](https://github.com/tenequm/pond/issues/171) - twelve harnesses | ✅ [v0.15.1](https://github.com/tenequm/pond/releases/tag/v0.15.1) |\n| 9 | Lance 10: count pushdown, date-filter zonemaps, faster index commits. [#145](https://github.com/tenequm/pond/issues/145) | ⏭ Next |\n| 10 | `pond erase`: the one sanctioned deletion. [#45](https://github.com/tenequm/pond/issues/45) | ⏭ Next |\n| 11 | Remote reads as fast as local reads. [#165](https://github.com/tenequm/pond/issues/165) | ⏳ Later |\n| 12 | Namespaces: keep work and personal sessions apart. [#166](https://github.com/tenequm/pond/issues/166) | ⏳ Later |\n| 13 | Redaction on copy, export, and resume. Never at ingest. [#167](https://github.com/tenequm/pond/issues/167) | ⏳ Later |\n| | A capture daemon. pond reads what your harness already writes. | ❌ Not planned |\n| | A hosted service that owns your bucket. Your storage stays yours. | ❌ Not planned |\n| | Summaries or pruning of stored sessions. pond keeps the sessions. | ❌ Not planned |\n\nStep 7 was based on data. Over 63 days, agents ran 1,126 searches against this archive. BM25 found the answer 61% of the time. Vector found it 37% of the time. Read [the measurement](https://github.com/tenequm/pond/blob/main/docs/researches/2608-21-semantic-vs-fts-usage-eval/README.md).\n\nThis is a direction, not a contract. The order changes when the data changes.\n\n## References\n\nThe upstream schemas that shaped pond's canonical model are documented in [`docs/references/`](https://github.com/tenequm/pond/blob/main/docs/references/) (source URLs + why each matters; the vendored code itself is not redistributed). Real session captures live under `packages/pond/tests/fixtures/adapter/`.\n\n| Source | Why it matters |\n|--------|----------------|\n| [Effect-TS/effect](https://github.com/Effect-TS/effect) | Effect v4 Prompt/Response Part unions. Pond's canonical types copy this shape. |\n| [sst/opencode](https://github.com/sst/opencode) | Effect Schema canonical Part union; SDK types; storage schema. |\n| [kilo-org/kilocode](https://github.com/kilo-org/kilocode) | OpenCode fork. Adds `editorContext`, plan-followup, kilocode-specific events. |\n| [badlogic/pi-mono](https://github.com/badlogic/pi-mono) | pi-coding-agent leaf-cursor branching and cross-provider conformance test matrix. |\n| [open-telemetry/semantic-conventions-genai](https://github.com/open-telemetry/semantic-conventions) | GenAI semantic conventions. Inspiration for shape overlap; pond does not derive from OTel. |\n| `packages/pond/tests/fixtures/adapter/` | Session samples for thirteen source harnesses (claude_ai_export, claude_code, claude_desktop_app, claude_managed_agents, codex_cli, grok-build, hermes, letta-code, nanoclaw, oh-my-pi, openclaw, opencode, pi-coding-agent; real captures except the synthetic hermes `state.db` and the generated oh-my-pi corpus). Drives adapter design and serves as adapter test fixtures. |\n\n## Contributing\n\nIssues and pull requests are welcome. The most useful contributions right now:\n\n- An adapter for a harness pond does not read yet. The playbook is the [`add-adapter` skill](https://github.com/tenequm/pond/blob/main/.agents/skills/add-adapter/SKILL.md); the PR expectations are in [CONTRIBUTING.md](https://github.com/tenequm/pond/blob/main/CONTRIBUTING.md).\n- Spec feedback on [`docs/spec.md`](https://github.com/tenequm/pond/blob/main/docs/spec.md).\n- Pointers to additional reference schemas or session samples worth documenting under `docs/references/`.\n- Bug reports against the v1 surface (CLI verbs, wire ops, schema mismatches, OCC behavior, object-store backends).\n\nFor something bigger, the [roadmap](#roadmap) is the list. Comment on the issue before you start, so we agree on the scope. For other larger changes, open an issue first to discuss the direction. For security issues, see [SECURITY.md](https://github.com/tenequm/pond/blob/main/.github/SECURITY.md).\n\nQuestions or feedback? Start a [GitHub Discussion](https://github.com/tenequm/pond/discussions), or DM me on [Telegram](https://t.me/tenequm) or [X](https://x.com/opwizardx) - I answer personally.\n\n## Links\n\n- Docs: [pond.locker](https://pond.locker/)\n- Crate: [pond-db on crates.io](https://crates.io/crates/pond-db)\n- MCP Registry name: `mcp-name: io.github.tenequm/pond`\n\n## License\n\n[Apache-2.0](https://github.com/tenequm/pond/blob/main/LICENSE) (c) 2026 tenequm\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt","priceCents","trialDays",0,"iconType","image","upstream",{"_106":107,"_108":109,"_110":111,"_112":113,"_114":115,"_116":117},"requiredSecrets",[],"tags",[95,96,97,98,99,100,101,102,103,104,105],"assets",[],"stats",{"_93":80,"_94":80},"ratingAvg","reviewCount","ai-agents","claude-code","codex","fts","lance","memory","opencode","pi-coding-agent","rust","session-logs","vector-search","topics",[95,96,97,98,99,123,124,100,101,102,103,104,105],"license","Apache-2.0","language","Rust","registryName","io.github.tenequm/pond","homepage","https://pond.locker/","packages",[118],{"_119":120,"_121":122,"_70":71},"registryType","cargo","identifier","pond-db","mcp","mcp-server","ENV",{"_127":128,"_129":130,"_131":132},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272"]