Endpoints: 28,729MCP servers: 18,413Payout addresses: 2,071Paid calls: 1,551Letters: 14Defects: 1,323counted 3 min ago
teppi

Server definition

Hash
sha256:1fa38c735764c0a7e170fb600ceec8f259b05bcfff526df8d5ed9f1cd4f45dee
What it is
What a remote MCP server returned when asked what it offers: 5 tools

The blob, as servednamed by its sha256

{ "instructions": "Project long-term memory. Call haki_context before planning or editing code, haki_capture at the end of a task for decisions, conventions and resolved errors. Only memorize durable content, never secrets. Call haki_correct when the user reports that a fact recalled by haki_context is wrong.", "tools": [ { "description": "Memorize a durable project fact: technical decision, convention,\n resolved error. Call at the END of a task. Never memorize secrets,\n tokens or ephemeral data. Consolidation is synchronous in dev\n (HAKI_MCP_AUTOCONSOLIDATE), so the fact is recallable immediately.\n Content larger than MAX_JSON_BYTES (256 KiB, same bound as POST\n /v1/capture) is rejected with a typed error.", "inputSchema": { "properties": { "content": { "title": "Content", "type": "string" }, "kind": { "default": "agent.observation", "title": "Kind", "type": "string" } }, "required": [ "content" ], "title": "haki_captureArguments", "type": "object" }, "name": "haki_capture", "outputSchema": { "additionalProperties": true, "title": "haki_captureDictOutput", "type": "object" } }, { "description": "Recall the project's memory (decisions, conventions, preferences)\n relevant to a task. Call BEFORE planning or editing code. Returns a\n ready-to-inject block with dates and sources, the trace_id (inspectable\n via haki_inspect), and an explicit `status` (\"ok\"/\"degraded\"/\"failed\") —\n never treat a degraded or failed response as \"no known facts\".\n\n If the block does not hold what was needed, call this tool again with\n the SAME query and `exclude_ids` set to the ids already received — it\n serves the next page of the same ranked list. Do NOT rewrite the query\n with what was just read: measured on this project's bench, that finds\n the missing turn less often than asking unchanged.", "inputSchema": { "properties": { "budget_tokens": { "default": 3000, "title": "Budget Tokens", "type": "integer" }, "exclude_ids": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "title": "Exclude Ids" }, "query": { "title": "Query", "type": "string" } }, "required": [ "query" ], "title": "haki_contextArguments", "type": "object" }, "name": "haki_context", "outputSchema": { "additionalProperties": true, "title": "haki_contextDictOutput", "type": "object" } }, { "description": "Correct the memory from inside the conversation (M10). rating='incorrect'\n with a fact_id moves that fact to 'disputed': haki_context will never\n recall it again. rating='useful'/'irrelevant' logs feedback on a recall\n without changing the fact. Exactly one target required: fact_id (one\n precise fact, usually seen via haki_context/haki_inspect) OR trace_id\n (one whole haki_context call). Same mechanism as POST /v1/feedback\n (app.ledger.submit_feedback): identical effect whatever the call path.\n Scope resolved by API key (see _resolve_scope) or, self-hosted without\n a key, by HAKI_MCP_PROJECT_ID.", "inputSchema": { "properties": { "comment": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "title": "Comment" }, "fact_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "title": "Fact Id" }, "rating": { "title": "Rating", "type": "string" }, "trace_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "title": "Trace Id" } }, "required": [ "rating" ], "title": "haki_correctArguments", "type": "object" }, "name": "haki_correct", "outputSchema": { "additionalProperties": true, "title": "haki_correctDictOutput", "type": "object" } }, { "description": "Forget the configured subject's memory in this project.\n mode='disable' (reversible, facts move to disabled) or 'delete' (real\n erasure: facts, embeddings, events, traces). Returns the erasure receipt\n (forget_id) and the counters of what was done.", "inputSchema": { "properties": { "mode": { "default": "disable", "title": "Mode", "type": "string" } }, "title": "haki_forgetArguments", "type": "object" }, "name": "haki_forget", "outputSchema": { "additionalProperties": true, "title": "haki_forgetDictOutput", "type": "object" } }, { "description": "Inspect the trace of a haki_context call: which facts were included,\n excluded or blocked, and why (reason_code). Provenance proof of the\n served memory.", "inputSchema": { "properties": { "trace_id": { "title": "Trace Id", "type": "string" } }, "required": [ "trace_id" ], "title": "haki_inspectArguments", "type": "object" }, "name": "haki_inspect", "outputSchema": { "additionalProperties": true, "title": "haki_inspectDictOutput", "type": "object" } } ] }
Verify it yourselfcurl -s https://api.teppi.xyz/v1/evidence/sha256:1fa38c735764c0a7e170fb600ceec8f259b05bcfff526df8d5ed9f1cd4f45dee | sha256sum