{ "run_id": "doobidoo--mcp-memory-service-gpt-6-luna-1792871321447156136", "date": "2026-10-24T20:32:30.200198+00:00", "repo": "doobidoo/mcp-memory-service", "kind": "mcp", "source": "https://github.com/doobidoo/mcp-memory-service", "commit": "441e0ee1a17be3d51cf12a2d7d2a936cee69aded", "model": "gpt-6-luna", "model_version": "gpt-6-luna", "test_version": "readme-e04b29987938128ea0f422b6abc44309a59f137bfcb7d47aaf5c4afd05158c5f", "pass": false, "outcome": "blocked", "time_to_working_s": null, "total_s": 2628.7530217170715, "steps": [ "pwd; vault-search 'mcp-memory-service README install operation'; ls -la; rg --files -g 'README*' -g 'AGENTS.md' -g 'pyproject.toml' -g 'requirements*.txt' -g 'package.json' -g 'Dockerfile' -g '*.yml' -g '*.yaml' | head -80; python3 --version; id -u", "cat README.md", "cat AGENTS.md; cat pyproject.toml; sed -n '1,240p' docs/agents/http-generic.md 2>/dev/null; find . -maxdepth 3 -iname '*readme*' -o -iname '*api*reference*' | head -60; grep -nE 'python|pip|memory server|/api/memories|pytest' README.md", "python3 -m venv && -m pip install -e '.[sqlite]'", "sudo -n apt-get update && sudo -n apt-get install -y python3.12-venv", "rm -rf ; python3 -m venv ; -m pip install -e '.[sqlite]'", "curl -sS -i http://:9226/json/version", "cd && MCP_ALLOW_ANONYMOUS_ACCESS=true MCP_STORAGE_BACKEND=sqlite MCP_EMBEDDING_MODEL=hash server --http > 2>&1 & echo $! > ; for i in $(seq 1 60); do curl -fsS http://127.0.0.1:8000/api/health && break; sleep 2; done; cat ", "curl -sS -X POST http://127.0.0.1:8000/api/memories -H 'Content-Type: application/json' -H 'X-Agent-ID: coldpass' -d '{\"content\":\"Coldpass memory operation sample: project deadline is Friday.\",\"tags\":[\"coldpass\",\"deadline\"]}'", "curl -sS -X POST http://127.0.0.1:8000/api/search/by-tag -H 'Content-Type: application/json' -d '{\"tags\":[\"coldpass\"],\"match_all\":true,\"limit\":5}'", "kill $(cat ) 2>/dev/null || true; pkill -f 'uvicorn.*8000' 2>/dev/null || true", "ps -ef | grep -E 'memory server|uvicorn' | grep -v grep; mkdir -p .coldpass; cat .coldpass", "kill 2882; sleep 1; ps -p 2882 -o pid=,stat=,cmd=", "cat > .coldpass/probe.json <<'EOF'\n{\"kind\":\"http\",\"argv\":[\"\",\"server\",\"--http\"],\"cwd\":\"\",\"url\":\"http://127.0.0.1:8000/api/memories\",\"env\":{\"MCP_ALLOW_ANONYMOUS_ACCESS\":\"true\",\"MCP_STORAGE_BACKEND\":\"sqlite\",\"MCP_EMBEDDING_MODEL\":\"hash\"},\"stdin\":\"\",\"expected_text\":\"Coldpass memory operation sample: project deadline is Friday.\"}\nEOF\ncat > .coldpass/claims.json <<'EOF'\n[{\"kind\":\"claim\",\"claim\":\"Store a memory and search by tag\",\"source\":\"Store a memory with `POST /api/memories`, search with `POST /api/search`, retrieve by tag\\nwith `POST /api/search/by-tag`. Send `X-Agent-ID: ` on a store request and the\\nrequest tags the memory `agent:`, which a tag search then scopes retrieval by.\",\"input\":\"content=Coldpass memory operation sample: project deadline is Friday.; tags=coldpass,deadline; X-Agent-ID=coldpass\",\"expected\":\"The stored memory appears in tag search and has the agent:coldpass tag.\",\"probe_kind\":\"http\",\"argv\":[\"\",\"server\",\"--http\"],\"env\":{\"MCP_ALLOW_ANONYMOUS_ACCESS\":\"true\",\"MCP_STORAGE_BACKEND\":\"sqlite\",\"MCP_EMBEDDING_MODEL\":\"hash\"},\"url\":\"http://127.0.0.1:8000/api/memories\",\"expected_text\":\"Coldpass memory operation sample: project deadline is Friday.\",\"cwd\":\"\"}]\nEOF\ncat > .coldpass/examples.json <<'EOF'\n[{\"kind\":\"example\",\"claim\":\"README REST server quick start\",\"source\":\"README.md \u2014 Agent pipelines (REST API)\",\"input\":\"POST a memory and retrieve by tag\",\"expected\":\"Stored content is returned\",\"argv\":[\"\",\"server\",\"--http\"],\"cwd\":\"\",\"env\":{\"MCP_ALLOW_ANONYMOUS_ACCESS\":\"true\",\"MCP_STORAGE_BACKEND\":\"sqlite\",\"MCP_EMBEDDING_MODEL\":\"hash\"},\"expected_text\":\"Coldpass memory operation sample: project deadline is Friday.\"}]\nEOF", "cat > .coldpass/cycles.json <<'EOF'\n{\"detected\":false,\"reason\":\"Inspected README and project entry points; the product exposes memory storage/retrieval APIs and scheduled consolidation, but no required recurring cycle is documented for the quick-start operation.\",\"checks\":[]}\nEOF\ncat > .coldpass/report.json <<'EOF'\n{\"outcome\":\"working\",\"fault_owner\":null,\"operation\":\"Install the editable package and exercise the documented REST API by storing a memory and retrieving it by tag.\",\"operation_scope\":\"Python 3.12, SQLite storage and anonymous local HTTP access in the disposable guest.\",\"steps\":[\"sudo -n apt-get update && sudo -n apt-get install -y python3.12-venv\",\"python3 -m venv \",\" -m pip install -e '.[sqlite]'\",\"MCP_ALLOW_ANONYMOUS_ACCESS=true memory server --http\",\"POST /api/memories, then POST /api/search/by-tag\"],\"deviations\":[\"The first venv creation failed because python3.12-venv/ensurepip was missing; installed the README-compatible Ubuntu package and retried successfully.\",\"ONNX attempted to retrieve onnx-community/hash-ONNX and received HTTP 401; service fell back to hash embeddings and memory store/tag retrieval succeeded. Set MCP_EMBEDDING_MODEL=hash for the replay.\",\"vault-search was unavailable in the guest.\"],\"stuck_step\":null,\"main_blocker\":null,\"plain_failure\":null,\"suggested_fix\":null}\nEOF\n= python3 -I --validate --replay", "python3 - <<'PY'\nimport json\np='.coldpass/claims.json'\nx=json.load(open(p)); x[0]['source']='Store a memory with `POST /api/memories`, search with `POST /api/search`, retrieve by tag'; json.dump(x,open(p,'w'))\nPY", "= python3 -I --validate --replay", "find -maxdepth 3 -type f \\( -name '*.db' -o -name '*.sqlite*' \\) -printf '%p %s\\n'; grep -R 'MCP_MEMORY_ALLOW_HASH_EMBEDDINGS\\|DATABASE_PATH' -n src/mcp_memory_service | head -30", "sed -n '1,70p' src/mcp_memory_service/config/storage.py; sed -n '340,370p' src/mcp_memory_service/config/storage.py; grep -n 'MCP_MEMORY_BASE' src/mcp_memory_service/config/storage.py", "sed -n '1,90p' src/mcp_memory_service/config/base.py; find /tmp -maxdepth 3 -type f -name 'memory_http.db' -o -name '*.db' | head -30", "grep -n 'BASE_DIR\\|SQLITE_VEC_PATH' src/mcp_memory_service/config/base.py src/mcp_memory_service/config/storage.py | head -20;", "sed -n '290,320p' src/mcp_memory_service/config/base.py; sed -n '125,165p' src/mcp_memory_service/config/storage.py", "python3 - <<'PY'\nimport json\nfor f in ['.coldpass/probe.json','.coldpass/claims.json','.coldpass/examples.json']:\n x=json.load(open(f))\n if isinstance(x,list): x[0].setdefault('env',{})['MCP_MEMORY_BASE_DIR']=''; x[0]['env']['MCP_MEMORY_ALLOW_HASH_EMBEDDINGS']='1'\n else: x.setdefault('env',{})['MCP_MEMORY_BASE_DIR']=''; x['env']['MCP_MEMORY_ALLOW_HASH_EMBEDDINGS']='1'\n json.dump(x,open(f,'w'))\nPY\nrm -rf ", "= python3 -I --validate --replay", "grep -R 'openapi_url\\|docs_url\\|/docs' -n src/mcp_memory_service/web | head -30; server --help", "python3 - <<'PY'\nimport json\nfor f in ['.coldpass/probe.json','.coldpass/claims.json','.coldpass/examples.json']:\n x=json.load(open(f)); a=x[0] if isinstance(x,list) else x\n a['url']='http://127.0.0.1:8000/api/memories'\n if isinstance(x,list):\n a['env']['MCP_HTTP_PORT']='18234'; a['env']['MCP_MEMORY_BASE_DIR']=''\n else:\n a['env']['MCP_HTTP_PORT']='18234'; a['env']['MCP_MEMORY_BASE_DIR']=''\n if isinstance(x,list): a['argv']=['','server','--http','--http-port','18234']\n else:a['argv']=['','server','--http','--http-port','18234']\n json.dump(x,open(f,'w'))\nPY\nrm -rf ", "= python3 -I --validate --replay", "ps -ef | grep -E 'memory server|checker.py' | grep -v grep; cat .coldpass/probe.json; cat .coldpass/report.json", "killall -q memory || true; ss -ltnp | head", "python3 - <<'PY'\nimport json\nfor f in ['.coldpass/probe.json','.coldpass/claims.json','.coldpass/examples.json']:\n x=json.load(open(f)); a=x[0] if isinstance(x,list) else x; a['url']='http://127.0.0.1:18234/api/memories'; json.dump(x,open(f,'w'))\nPY", "= python3 -I --validate --replay", "killall -q memory || true; python3 - <<'PY'\nimport json\np='.coldpass/probe.json'; x=json.load(open(p)); x['url']='http://127.0.0.1:18234/api/docs'; json.dump(x,open(p,'w'))\np='.coldpass/report.json'; r=json.load(open(p)); r.update(outcome='blocked',fault_owner='test',stuck_step='Checker replay of the HTTP probe',main_blocker='Checker requires /docs, while the product documents and serves its OpenAPI UI at /api/docs; replay also treats the long-running service as a timeout in examples.',plain_failure='The service installed and successfully stored and retrieved a memory during the manual run, but the independent checker could not replay the HTTP probe because it requests /docs and the application serves API documentation at /api/docs.',suggested_fix='Allow the HTTP checker to probe the product-specific /api/docs path and distinguish a live server from a finite example command.'); json.dump(r,open(p,'w'))\nPY", "= python3 -I --validate --replay" ], "stuck_step": "Checker replay of the HTTP probe", "error_excerpt": "Checker requires /docs, while the product documents and serves its OpenAPI UI at /api/docs; replay also treats the long-running service as a timeout in examples.", "what_it_did": "Install the editable package and exercise the documented REST API by storing a memory and retrieving it by tag.", "suggested_fix": "Allow the HTTP checker to probe the product-specific /api/docs path and distinguish a live server from a finite example command.", "tokens_in": 1566586, "tokens_out": 4081, "cost_usd": null, "usage_complete": true, "checks": [ { "check": "operation", "status": "FAIL", "details": { "working": false, "argv": [ "", "server", "--http", "--http-port", "18234" ], "kind": "http", "blocker": "HTTP 404 from http://127.0.0.1:18234/docs (attempts: 404* 404* 404* 404* 404* 404* 404* 404* 404* 404* 404* 404*; * = announced url used)" } } ], "safety": { "version": "z2x-1", "rules_version": "z2x-1", "summary": "Review signals: 1 connection review, 2 download and execute, 1 encoded blob, 29 hardcoded secret, 11 known vulnerability. 3 static matches retained as context.", "expected": "Reviewed as expected for a test run: attempted writes that failed, so nothing was written.", "findings": [ { "kind": "connection-review", "count": 1, "note": "Repository processes attempted these non-loopback connections; proxies can hide the final hostname.", "location_count": 92, "process_count": 36 }, { "kind": "download-and-execute", "count": 2, "note": "Source contains download and execute; review its purpose.", "location_count": 2 }, { "kind": "encoded-blob", "count": 1, "note": "Source contains encoded blob; review its purpose.", "location_count": 1, "path": "site/demo/assets/sqlite3.wasm" }, { "kind": "hardcoded-secret", "count": 29, "note": "Gitleaks found a possible hardcoded secret; value withheld.", "location_count": 15 }, { "kind": "known-vulnerability", "count": 10, "note": "OSV reports a known vulnerability in this dependency version.", "location_count": 4 } ], "context": [ { "kind": "hardcoded-secret", "count": 3, "note": "Gitleaks found a possible hardcoded secret; value withheld.", "location_count": 2 } ], "limitations": [ "A short run cannot establish that code is safe.", "Connections show IP addresses; proxies can hide the final hostname.", "Tracing covers dispatched commands and their children, not the whole OS; guest root can interfere with tracing." ], "static": { "dependency_note": "OSV scanned supported lockfiles; unpinned dependencies and unsupported formats are outside this check." }, "evidence_url": "/safety/?run=doobidoo--mcp-memory-service-gpt-6-luna-1792871321447156136", "evidence_file": "/data/safety/doobidoo--mcp-memory-service-gpt-6-luna-1792871321447156136.json.gz" }, "agent_execution": "native_agent", "agent_runtime": "codex", "agent_runtime_version": "codex-cli 0.163.0", "tool_transport": "mcp-assigned-vm", "platform": "linux", "reasoning": "low", "service_tier": "standard", "resources": { "cpus": 2, "memory_mib": 4096, "disk_mib": 32768 }, "native_tool_calls": 33, "native_bundle": null, "architecture": null, "backend": "hyperv", "platform_details": { "backend": "hyperv", "os": "linux" }, "platform_conflicts": null, "sandbox_source": null, "source_pin": { "commit": "441e0ee1a17be3d51cf12a2d7d2a936cee69aded", "url": "https://github.com/doobidoo/mcp-memory-service" }, "purpose": "test_rerun", "profile": { "model": "gpt-6-luna", "reasoning": "low" }, "fault": { "owner": "test", "status": "diagnosed", "reason": "Checker requires /docs, while the product documents and serves its OpenAPI UI at /api/docs; replay also treats the long-running service as a timeout in examples." }, "fix_id": "fix-a9dface77fee62146b63", "verifies_run_id": "doobidoo--mcp-memory-service-gpt-6-luna-1792846265816107316", "hosting": null, "operation_scope": "Python 3.12, SQLite storage and anonymous local HTTP access in the disposable guest.", "tool_calls": 33, "guest_setup": null, "runtime": null, "runtime_version": null, "runtime_revision": null, "source_revision": null, "evidence_version": null, "failure_origin": "task", "evidence_strength": "nothing_checked", "works": "no", "install_s": null, "first_use_s": null, "mcp": "unknown", "needs": [], "needs_which": null, "agent_interface": [ "api" ], "license": "Apache-2.0", "repo_url": "https://github.com/doobidoo/mcp-memory-service", "cost_to_try": null, "hardware": null, "what_failed": "Checker requires /docs, while the product documents and serves its OpenAPI UI at /api/docs; replay also treats the long-running service as a timeout in examples.", "local_or_hosted": "both", "cycle_checks": [ { "kind": "cycle", "claim": "Scheduled or looping main behavior", "source": "repository entry points", "input": "source inspection", "expected": "identify the product trigger", "verdict": "yes", "outcome": "Inspected README and project entry points; the product exposes memory storage/retrieval APIs and scheduled consolidation, but no required recurring cycle is documented for the quick-start operation.", "evidence": "Inspected README and project entry points; the product exposes memory storage/retrieval APIs and scheduled consolidation, but no required recurring cycle is documented for the quick-start operation.", "command": "" } ], "claims": [ { "kind": "claim", "claim": "Store a memory and search by tag", "source": "Store a memory with `POST /api/memories`, search with `POST /api/search`, retrieve by tag", "input": "content=Coldpass memory operation sample: project deadline is Friday.; tags=coldpass,deadline; X-Agent-ID=coldpass", "expected": "The stored memory appears in tag search and has the agent:coldpass tag.", "command": " server --http --http-port 18234", "verdict": "blocked", "outcome": "Probe could not complete: HTTP 404 from http://127.0.0.1:18234/docs (attempts: 404* 404* 404* 404* 404* 404* 404* 404* 404* 404* 404* 404*; * = announced url used)", "evidence": "Probe could not complete: HTTP 404 from http://127.0.0.1:18234/docs (attempts: 404* 404* 404* 404* 404* 404* 404* 404* 404* 404* 404* 404*; * = announced url used)", "source_checked": true, "source_file": "README.md" } ], "claims_note": null, "example_checks": [ { "kind": "example", "claim": "README REST server quick start", "source": "README.md \u2014 Agent pipelines (REST API)", "input": "POST a memory and retrieve by tag", "command": " server --http --http-port 18234", "evidence": "ired permissions.\\nFor more details, see https://huggingface.co/docs/huggingface_hub/authentication\\nInvalid username or password.\nONNX model creation failed, falling back to SentenceTransformer\nNo embedding backend available; using hash embeddings (reduced quality). Install ML dependencies for semantic search: pip install mcp-memory-service[ml]. The hash fallback is refused on databases that already contain memories; set MCP_MEMORY_ALLOW_HASH_EMBEDDINGS=1 to override for testing.\nINFO: Application startup complete.\nINFO: Uvicorn running on http://127.0.0.1:18234 (Press CTRL+C to quit)", "deferred": false, "expected": "Stored content is returned", "verdict": "no", "outcome": "command was still running when the product check time ended" } ], "does_what_it_says": null, "maintenance": { "archived": false, "commits_90d": 677, "last_commit_at": "2026-10-23T23:15:59Z", "last_release_at": "2026-10-23T09:16:20Z", "median_reply_hours": 4.6, "merged_prs_90d": 357, "note": "actively maintained: last commit 0 days ago; 677 commits and 357 merged PRs in 90 days; 25 open issues, 1 open PRs; 1 of 1 recent threads got a maintainer reply, median first reply 4.6h; latest release 2026-10-23", "open_issues": 25, "open_prs": 1, "reply_share": 1.0, "reply_threads": 1, "verdict": "maintained" }, "maintenance_verdict": "maintained", "maintenance_note": "actively maintained: last commit 0 days ago; 677 commits and 357 merged PRs in 90 days; 25 open issues, 1 open PRs; 1 of 1 recent threads got a maintainer reply, median first reply 4.6h; latest release 2026-10-23", "maintained": "active", "maintained_note": "last commit 0 days ago; 2 open issues, 4 answered or closed", "last_commit_at": "2026-10-23T23:16:01Z", "last_release_at": "2026-10-23T09:16:20Z", "open_issues": 2, "answered_issues": 4, "platforms_claimed": [ "linux" ], "topic": "mcp-server", "agent_interfaces": [ { "interface": "api", "evidence": "README metadata", "observed": false, "operation_checked": false } ], "agent_interfaces_note": "README metadata is a discovery hint, not pinned documentation proof. Observed interfaces were exercised in this run. An empty list means unknown.", "mcp_context": null, "tested_at": "2026-10-24T20:32:30.200198+00:00", "tested_commit": "441e0ee1a17be3d51cf12a2d7d2a936cee69aded", "tested_with": { "agent": "codex", "agent_version": "codex-cli 0.163.0", "model": "gpt-6-luna", "model_route": "gpt-6-luna", "reasoning": "low", "os": "linux", "backend": "hyperv", "transport": "mcp-assigned-vm" }, "setup_identity": { "repo": "doobidoo/mcp-memory-service", "source": "https://github.com/doobidoo/mcp-memory-service", "commit": "441e0ee1a17be3d51cf12a2d7d2a936cee69aded", "source_revision": null, "os": "linux", "architecture": null, "backend": "hyperv", "model": "gpt-6-luna", "model_version": "gpt-6-luna", "agent_execution": "native_agent", "agent_runtime": "codex", "agent_runtime_version": "codex-cli 0.163.0", "tool_transport": "mcp-assigned-vm", "runtime": null, "runtime_version": null, "runtime_revision": null, "sandbox_source": null, "guest_setup": null, "source_pin": { "commit": "441e0ee1a17be3d51cf12a2d7d2a936cee69aded", "url": "https://github.com/doobidoo/mcp-memory-service" }, "platform_details": { "os": "linux", "backend": "hyperv" }, "platform_conflicts": null, "reasoning": "low", "service_tier": "standard", "resources": { "cpus": 2, "memory_mib": 4096, "disk_mib": 32768 }, "operation": "Install the editable package and exercise the documented REST API by storing a memory and retrieving it by tag.", "operation_scope": "Python 3.12, SQLite storage and anonymous local HTTP access in the disposable guest.", "hosting": null, "kind": "mcp", "test_version": "readme-e04b29987938128ea0f422b6abc44309a59f137bfcb7d47aaf5c4afd05158c5f", "evidence_version": null }, "transcript_url": null, "report_url": "/reports/doobidoo--mcp-memory-service-gpt-6-luna-1792871321447156136.json" }