[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M69BGVHJJFCNZMGM21FNWMFS":3,"skill-same:01M69BGVHJJFCNZMGM21FNWMFS":49},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M69BGVHJJFCNZMGM21FNWMFS","context-intelligence-session-reconstruction",null,"Use when local Amplifier session files are missing, broken, or incomplete — rebuilds events.jsonl, transcript.jsonl, and metadata.json from the context-intelligence graph server.","microsoft\u002Famplifier-bundle-context-intelligence","microsoft","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6154722?v=4","01M698VYDXZ6JW5WH3CRV650R8","Microsoft","01M69BGBVCCWXG1MNNF1E8TCWN",5,"MIT","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fcontext-intelligence-session-reconstruction\u002FSKILL.md",[],[19],"context intelligence session reconstruction","2026-11-10 10:12:30",false,"---\nname: context-intelligence-session-reconstruction\nversion: 1.1.0\ndescription: Use when local Amplifier session files are missing, broken, or incomplete — rebuilds events.jsonl, transcript.jsonl, and metadata.json from the context-intelligence graph server.\nlicense: MIT\n# Fallback-path skill: reached deliberately, by name, when local session files are\n# already known to be missing or broken — never a candidate the model should pick\n# opportunistically off the always-visible catalog. Hidden from the \"Available\n# skills\" index; still loads exactly as before via\n# load_skill(skill_name=\"context-intelligence-session-reconstruction\").\ndisable-model-invocation: true\n---\n\n# Context Intelligence Session Reconstruction\n\nLocal Amplifier session files (`events.jsonl`, `transcript.jsonl`, `metadata.json`) can be reconstructed from the context-intelligence Neo4j graph server when the local files are missing, broken, or incomplete. This skill covers when and how to run the reconstruction tool safely.\n\n---\n\n## When to Use\n\nUse reconstruction when local session files are absent or unusable but the graph server holds the session data:\n\n- **Missing files**: Sessions exist in the graph but have no local files — for example, after a disk wipe, new machine setup, or accidental deletion\n- **Broken resume**: `amplifier resume` shows \"unnamed\", \"unknown\", or empty session names — reconstructing `metadata.json` populates bundle, model, and session name fields\n- **Unknown bundles**: Sessions that show no bundle information in the resume list — reconstruction extracts bundle info from the graph's `session_start` blob\n\n---\n\n## When NOT to Use\n\n- **Server unreachable**: If the graph server is not reachable, reconstruction cannot proceed. Verify server reachability first with `python scripts\u002Fcontext-intelligence.py status`. If the server is down, fall back to `session-navigator` for local file navigation.\n- **Pre-hook sessions**: Sessions that predate the `hook-context-intelligence` hook have no data in the graph. Reconstruction cannot create files for these sessions — there is no source data to read from.\n- **Real-time data**: Reconstruction reads a snapshot of what is stored in the graph. It is not a live stream. Do not use it to monitor an active session or capture events as they happen.\n\n---\n\n## Prerequisites\n\n**1. Graph server must be reachable**\n\nVerify with the status subcommand before running reconstruction:\n\n```bash\npython scripts\u002Fcontext-intelligence.py status\n```\n\nA healthy server responds with `\"status\": \"ok\"`. If the server is unreachable, this command will report the error. Do not proceed with reconstruction until status is `ok`.\n\n**2. API key must be configured**\n\nThree methods — use whichever fits your environment:\n\n| Method | How |\n|--------|-----|\n| Environment variable | `export AMPLIFIER_CONTEXT_INTELLIGENCE_API_KEY=\u003Ckey>` |\n| CLI flag | Pass `--api-key \u003Ckey>` to any subcommand |\n| Settings file | Add `context_intelligence_api_key: \u003Ckey>` to `~\u002F.amplifier\u002Fsettings.yaml` |\n\nThe CLI resolves in this order: flag → env var → `~\u002F.amplifier\u002Fsettings.yaml`.\n\n**3. Sessions must have been captured by the hook**\n\nOnly sessions that were active while `hook-context-intelligence` was configured will have data in the graph. Pre-hook sessions cannot be reconstructed.\n\n---\n\n## Usage Patterns\n\nAll patterns use:\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct [OPTIONS]\n```\n\n### Fix broken resume list (fastest, recommended first)\n\nReconstructs only `metadata.json` without touching `events.jsonl` or `transcript.jsonl`. Use this when `amplifier resume` shows unnamed or unknown sessions.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --metadata-only\n```\n\n### Preview what would be written (dry run)\n\nShows what files would be created or updated without writing anything. Add `--verbose` for detailed per-session output.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --dry-run --verbose\n```\n\n### Reconstruct a single session\n\nLimits reconstruction to one session by ID or prefix. Use when you know exactly which session needs repair.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --session \u003Csession-id-or-prefix> --force\n```\n\nThe `--force` flag overwrites any existing files for that session. Omit `--force` to skip files that already exist.\n\n### Full reconstruction — skip existing files\n\nReconstructs all sessions in the project, skipping sessions where files already exist. Safe to run repeatedly.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct\n```\n\n### Full reconstruction — overwrite all files\n\nReconstructs all sessions and overwrites existing files. Use only when explicitly requested or when existing files are known to be corrupt.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --force\n```\n\n### Reconstruct with raw LLM data (large output)\n\nThe `--resolve-blobs` flag inlines full blob payloads into `events.jsonl`, including complete LLM request\u002Fresponse bodies. This produces significantly larger files (~19MB for a typical long session with many LLM turns).\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --resolve-blobs\n```\n\n> Use `--resolve-blobs` only when you need access to raw LLM input\u002Foutput. For most use cases — resume list repair, event counts, delegation trees — reconstructing without blobs is sufficient and much faster.\n\n---\n\n## Verification\n\nAfter reconstruction, verify the output files were written correctly:\n\n**Check line counts for events and transcript:**\n\n```bash\nCONTEXT_INTELLIGENCE_ROOT=\"${AMPLIFIER_CONTEXT_INTELLIGENCE_BASE_PATH:-$HOME\u002F.amplifier\u002Fprojects}\"\nwc -l \"$CONTEXT_INTELLIGENCE_ROOT\"\u002F{workspace}\u002Fsessions\u002F{session-id}\u002Fcontext-intelligence\u002Fevents.jsonl\nwc -l \"$CONTEXT_INTELLIGENCE_ROOT\"\u002F{workspace}\u002Fsessions\u002F{session-id}\u002Fcontext-intelligence\u002Ftranscript.jsonl\n```\n\nA reconstructed `events.jsonl` should have at least a few lines (session start, prompts, tool calls). Zero lines indicates the session had no recoverable events.\n\n**Validate metadata JSON is well-formed:**\n\n```bash\nCONTEXT_INTELLIGENCE_ROOT=\"${AMPLIFIER_CONTEXT_INTELLIGENCE_BASE_PATH:-$HOME\u002F.amplifier\u002Fprojects}\"\ncat \"$CONTEXT_INTELLIGENCE_ROOT\"\u002F{workspace}\u002Fsessions\u002F{session-id}\u002Fcontext-intelligence\u002Fmetadata.json | python3 -m json.tool\n```\n\nThis confirms the file is valid JSON. Check that `bundle`, `model`, and `session_id` fields are populated.\n\n**Confirm sessions appear in the resume list:**\n\n```bash\namplifier session list\n```\n\nSessions with successfully reconstructed `metadata.json` should now appear with names, bundle info, and model rather than \"unknown\" or empty fields.\n\n---\n\n## Known Limitations\n\n- **Streaming events not recoverable (~39%)**: `content_block:start` and `content_block:end` events are not stored in the graph — they account for approximately 39% of hook-logging events. Reconstructed `events.jsonl` files will be missing these events. This does not affect session content, only streaming telemetry.\n\n- **Delegate events may be incomplete**: `delegate:agent_spawned` and `delegate:agent_completed` events depend on `data_delegate_*` graph properties that may be null for sessions that were in progress when captured. These events may appear with partial data in the reconstructed files.\n\n- **Session names are approximations**: Session names are generated at runtime by `hooks-session-naming` from the first user prompt. That output is not stored in the graph. Reconstruction generates names from the same source (the first prompt), but the exact phrasing may differ from the original session name.\n\n- **Pre-hook sessions have no graph data**: Sessions that predate the `hook-context-intelligence` hook have no entries in the graph. Reconstruction will produce empty output for these sessions — there is no data to read. Use `session-navigator` to work with pre-hook sessions directly from local JSONL files if they still exist on disk.\n","# Context Intelligence Session Reconstruction\n\nLocal Amplifier session files (`events.jsonl`, `transcript.jsonl`, `metadata.json`) can be reconstructed from the context-intelligence Neo4j graph server when the local files are missing, broken, or incomplete. This skill covers when and how to run the reconstruction tool safely.\n\n---\n\n## When to Use\n\nUse reconstruction when local session files are absent or unusable but the graph server holds the session data:\n\n- **Missing files**: Sessions exist in the graph but have no local files — for example, after a disk wipe, new machine setup, or accidental deletion\n- **Broken resume**: `amplifier resume` shows \"unnamed\", \"unknown\", or empty session names — reconstructing `metadata.json` populates bundle, model, and session name fields\n- **Unknown bundles**: Sessions that show no bundle information in the resume list — reconstruction extracts bundle info from the graph's `session_start` blob\n\n---\n\n## When NOT to Use\n\n- **Server unreachable**: If the graph server is not reachable, reconstruction cannot proceed. Verify server reachability first with `python scripts\u002Fcontext-intelligence.py status`. If the server is down, fall back to `session-navigator` for local file navigation.\n- **Pre-hook sessions**: Sessions that predate the `hook-context-intelligence` hook have no data in the graph. Reconstruction cannot create files for these sessions — there is no source data to read from.\n- **Real-time data**: Reconstruction reads a snapshot of what is stored in the graph. It is not a live stream. Do not use it to monitor an active session or capture events as they happen.\n\n---\n\n## Prerequisites\n\n**1. Graph server must be reachable**\n\nVerify with the status subcommand before running reconstruction:\n\n```bash\npython scripts\u002Fcontext-intelligence.py status\n```\n\nA healthy server responds with `\"status\": \"ok\"`. If the server is unreachable, this command will report the error. Do not proceed with reconstruction until status is `ok`.\n\n**2. API key must be configured**\n\nThree methods — use whichever fits your environment:\n\n| Method | How |\n|--------|-----|\n| Environment variable | `export AMPLIFIER_CONTEXT_INTELLIGENCE_API_KEY=\u003Ckey>` |\n| CLI flag | Pass `--api-key \u003Ckey>` to any subcommand |\n| Settings file | Add `context_intelligence_api_key: \u003Ckey>` to `~\u002F.amplifier\u002Fsettings.yaml` |\n\nThe CLI resolves in this order: flag → env var → `~\u002F.amplifier\u002Fsettings.yaml`.\n\n**3. Sessions must have been captured by the hook**\n\nOnly sessions that were active while `hook-context-intelligence` was configured will have data in the graph. Pre-hook sessions cannot be reconstructed.\n\n---\n\n## Usage Patterns\n\nAll patterns use:\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct [OPTIONS]\n```\n\n### Fix broken resume list (fastest, recommended first)\n\nReconstructs only `metadata.json` without touching `events.jsonl` or `transcript.jsonl`. Use this when `amplifier resume` shows unnamed or unknown sessions.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --metadata-only\n```\n\n### Preview what would be written (dry run)\n\nShows what files would be created or updated without writing anything. Add `--verbose` for detailed per-session output.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --dry-run --verbose\n```\n\n### Reconstruct a single session\n\nLimits reconstruction to one session by ID or prefix. Use when you know exactly which session needs repair.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --session \u003Csession-id-or-prefix> --force\n```\n\nThe `--force` flag overwrites any existing files for that session. Omit `--force` to skip files that already exist.\n\n### Full reconstruction — skip existing files\n\nReconstructs all sessions in the project, skipping sessions where files already exist. Safe to run repeatedly.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct\n```\n\n### Full reconstruction — overwrite all files\n\nReconstructs all sessions and overwrites existing files. Use only when explicitly requested or when existing files are known to be corrupt.\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --force\n```\n\n### Reconstruct with raw LLM data (large output)\n\nThe `--resolve-blobs` flag inlines full blob payloads into `events.jsonl`, including complete LLM request\u002Fresponse bodies. This produces significantly larger files (~19MB for a typical long session with many LLM turns).\n\n```bash\npython scripts\u002Fcontext-intelligence.py reconstruct --resolve-blobs\n```\n\n> Use `--resolve-blobs` only when you need access to raw LLM input\u002Foutput. For most use cases — resume list repair, event counts, delegation trees — reconstructing without blobs is sufficient and much faster.\n\n---\n\n## Verification\n\nAfter reconstruction, verify the output files were written correctly:\n\n**Check line counts for events and transcript:**\n\n```bash\nCONTEXT_INTELLIGENCE_ROOT=\"${AMPLIFIER_CONTEXT_INTELLIGENCE_BASE_PATH:-$HOME\u002F.amplifier\u002Fprojects}\"\nwc -l \"$CONTEXT_INTELLIGENCE_ROOT\"\u002F{workspace}\u002Fsessions\u002F{session-id}\u002Fcontext-intelligence\u002Fevents.jsonl\nwc -l \"$CONTEXT_INTELLIGENCE_ROOT\"\u002F{workspace}\u002Fsessions\u002F{session-id}\u002Fcontext-intelligence\u002Ftranscript.jsonl\n```\n\nA reconstructed `events.jsonl` should have at least a few lines (session start, prompts, tool calls). Zero lines indicates the session had no recoverable events.\n\n**Validate metadata JSON is well-formed:**\n\n```bash\nCONTEXT_INTELLIGENCE_ROOT=\"${AMPLIFIER_CONTEXT_INTELLIGENCE_BASE_PATH:-$HOME\u002F.amplifier\u002Fprojects}\"\ncat \"$CONTEXT_INTELLIGENCE_ROOT\"\u002F{workspace}\u002Fsessions\u002F{session-id}\u002Fcontext-intelligence\u002Fmetadata.json | python3 -m json.tool\n```\n\nThis confirms the file is valid JSON. Check that `bundle`, `model`, and `session_id` fields are populated.\n\n**Confirm sessions appear in the resume list:**\n\n```bash\namplifier session list\n```\n\nSessions with successfully reconstructed `metadata.json` should now appear with names, bundle info, and model rather than \"unknown\" or empty fields.\n\n---\n\n## Known Limitations\n\n- **Streaming events not recoverable (~39%)**: `content_block:start` and `content_block:end` events are not stored in the graph — they account for approximately 39% of hook-logging events. Reconstructed `events.jsonl` files will be missing these events. This does not affect session content, only streaming telemetry.\n\n- **Delegate events may be incomplete**: `delegate:agent_spawned` and `delegate:agent_completed` events depend on `data_delegate_*` graph properties that may be null for sessions that were in progress when captured. These events may appear with partial data in the reconstructed files.\n\n- **Session names are approximations**: Session names are generated at runtime by `hooks-session-naming` from the first user prompt. That output is not stored in the graph. Reconstruction generates names from the same source (the first prompt), but the exact phrasing may differ from the original session name.\n\n- **Pre-hook sessions have no graph data**: Sessions that predate the `hook-context-intelligence` hook have no entries in the graph. Reconstruction will produce empty output for these sessions — there is no data to read. Use `session-navigator` to work with pre-hook sessions directly from local JSONL files if they still exist on disk.\n",{"name":5,"license":15,"version":25,"description":7,"disable-model-invocation":26},"1.1.0",true,"main","2a8d7bd1e022971c395ec38ed797bf75fcf7aff7","skills\u002Fcontext-intelligence-session-reconstruction",7,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence","https:\u002F\u002Fgithub.com\u002Fmicrosoft","https:\u002F\u002Fopensource.microsoft.com","Open source projects and samples from Microsoft","2026-11-07 04:52:31",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":26},[38,41,44],{"tool":39,"command":40},"npx","npx skills add microsoft\u002Famplifier-bundle-context-intelligence --skill context-intelligence-session-reconstruction",{"tool":42,"command":43},"bunx","bunx skills add microsoft\u002Famplifier-bundle-context-intelligence --skill context-intelligence-session-reconstruction",{"tool":45,"command":46},"pnpm","pnpm dlx skills add microsoft\u002Famplifier-bundle-context-intelligence --skill context-intelligence-session-reconstruction","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fcontext-intelligence-session-reconstruction\u002FSKILL.md",[],[50,67,76,93,101,117],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":61,"publishedAt":20},"01M69BGZVPJBKSZV5N0WMVGR3N","workflow-pattern-analysis","Use when asked how a workflow is failing, what separates a successful run from a failing one, which steps are the most common failure points, or what patterns recur across sessions — analyses failure\u002Fsuccess patterns and signals across many runs using context-intelligence session data.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fworkflow-pattern-analysis\u002FSKILL.md",[56],{"categoryNo":57,"categoryName":58,"taxonomyType":59,"taxonomyTypeDesc":60},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[62,63,64,65,66],"workflow pattern analysis","workflow","reasoning","run","sessions",{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":20},"01M69BGYAYC16BRSB0EGMCKMF5","transcript","Retrieve the prior user and assistant conversation verbatim from the current Context Intelligence capture, or named sessions. Use when a user asks to replay, quote, review, or act on a transcript.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Ftranscript\u002FSKILL.md",[],[69,74,66,75],"assistant","review",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":90,"publishedAt":20},"01M69BGWQGCEN13DHHAQGTQ609","context-intelligence-tool-design","Use when selecting a detection strategy and implementation primitive for a context-intelligence signal. Classifies signals as deterministic, probabilistic, llm-evaluated, or hybrid, and applies the cheapest-sufficient-capability principle.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fcontext-intelligence-tool-design\u002FSKILL.md",[82,85],{"categoryNo":83,"categoryName":84,"taxonomyType":59,"taxonomyTypeDesc":60},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":86,"categoryName":87,"taxonomyType":88,"taxonomyTypeDesc":89},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[91,92,64],"design","context intelligence tool design",{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":20},"01M69BGT14VQD1M3P12WZGTQCS","context-intelligence-session-navigation","Use when extracting session data directly from JSONL files — the fallback when the graph server is unavailable or outside graph-analyst.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fcontext-intelligence-session-navigation\u002FSKILL.md",[],[100],"context intelligence session navigation",{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":113,"publishedAt":20},"01M69BGRET24BR6PTYANVE57F4","context-intelligence-server-data-ops","Exact step order, wording, and the \"session details\" block format for the delete flows the server-data-ops agent drives — delete the current session, clean up every session from the current working directory, find-then-delete a session by description, and delete a session someone else created.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fcontext-intelligence-server-data-ops\u002FSKILL.md",[107,110],{"categoryNo":108,"categoryName":109,"taxonomyType":59,"taxonomyTypeDesc":60},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":111,"categoryName":112,"taxonomyType":88,"taxonomyTypeDesc":89},"01JBSEEDCAT000000000000033","数据分析师",[114,115,116],"context intelligence server data ops","delete","directory",{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":20},"01M69BGQ0B4G6VY99C6RGS8Z8T","context-intelligence-hill-climbing","Use when an investigation needs more than one step — track it in todo as a hill climb so progress and dead leads leave an audit trail. Governs HOW you track the climb, not how you query or extract — that stays in graph-query \u002F session-navigation.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-context-intelligence\u002Fblob\u002Fmain\u002Fskills\u002Fcontext-intelligence-hill-climbing\u002FSKILL.md",[],[124],"context intelligence hill climbing"]