[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DM9CSQVXGESK4SD0W0503T":3,"skill-same:01M7DM9CSQVXGESK4SD0W0503T":59},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":36,"commitSha":37,"skillDirectory":38,"forksCount":39,"repositoryHtmlUrl":40,"repositoryHomepageUrl":6,"ownerHtmlUrl":41,"ownerBlogUrl":42,"ownerDescription":43,"updatedAt":44,"install":45,"resources":58},"01M7DM9CSQVXGESK4SD0W0503T","agent-startup-and-context-ingestion",null,"Use when designing or auditing how an agent becomes useful after launch — AGENTS.md overlays, role files, skills, rig specs, workflow specs, startup checklists, refocus messages, \"rig context\" surface. Covers the 4 failure modes that make startup context fail (old rig spec misses current operating mode; current agents never told about new guidance; startup file as dumping ground; orchestrator transmits implementation without preserving product intent).","mvschwarz\u002Fopenrig-world","mvschwarz","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F171890339?v=4","01M7DM95A4N5936P8G2FTSPMY7","Mike Schwarz","01M7DM95AX80NY8HZA7XXDYAZR",16,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Fagents\u002Fshared\u002Fskills\u002Fcore\u002Fagent-startup-and-context-ingestion\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: agent-startup-and-context-ingestion\ndescription: Use when designing or auditing how an agent becomes useful after launch — AGENTS.md overlays, role files, skills, rig specs, workflow specs, startup checklists, refocus messages, \"rig context\" surface. Covers the 4 failure modes that make startup context fail (old rig spec misses current operating mode; current agents never told about new guidance; startup file as dumping ground; orchestrator transmits implementation without preserving product intent).\nmetadata:\n  cli_surfaces_referenced:\n    - context\n    - whoami\n  openrig:\n    stage: factory-approved\n    sibling_skills:\n      - claude-compaction-restore\n      - session-compaction-and-restore\n      - agent-starters\n      - session-source-fork\n      - seat-continuity-and-handover\n---\n\n# Agent Startup and Context Ingestion\n\nHow an agent becomes useful after launch: **AGENTS.md overlays, role\nfiles, skills, rig specs, workflow specs, startup checklists, refocus\nmessages**, and the current `rig context` retrieval and profile surface.\n\nIt is the current concrete startup path inside the broader\ncontext-engineering-and-retrieval primitive. Startup gets a seat into\nthe right initial shape; context engineering is the larger question of\nhow a seat gets the right context for the work it is doing right now.\n\n**Most coordination failures are not tool failures; they are context\nfailures.** Agents need to know their role, operating mode, coordination\nconvention, boundaries, and current product intent. If startup context\nis scattered or stale, agents execute the wrong thing very efficiently.\n\n## Use this when\n\n- Authoring a new agent's startup files (role \u002F culture \u002F startup-context)\n- Refreshing a seat that's been running on stale guidance\n- Auditing whether agents got the current operating mode (not just what's in old files)\n- Designing the orchestrator → next-agent context-transmission shape\n- Building a startup map for OpenRig-building rigs\n\n## Don't use this when\n\n- The agent is being created via Agent Starter — the starter's manifest carries startup context\n- The work is artifact-backed mental-model rebuild from a packet — that's `session-compaction-and-restore`\n- The intent is to ship reusable startup content as a skill\n\n## Failure modes (4)\n\n1. **A new agent starts from an old rig spec and misses the current operating mode.** Specs go stale; current state must be visible at startup, not just historical config.\n2. **Guidance is written to a file that future agents read, but current agents are never told.** File edits don't propagate to running sessions. Cultural rollout (broadcast + fleet-changes-feed) is needed alongside file edits.\n3. **A startup file becomes a dumping ground and loses the map-to-canonical-sources role.** Startup should point AT canonical sources; it shouldn't TRY to be one.\n4. **The orchestrator transmits implementation instructions without preserving product intent.** Instructions decay; intent travels.\n\n## Task-scoped startup\n\nRun `rig whoami --json`, then resolve `project.yaml -> mission.yaml -> active\nslice.yaml -> selected component or wave map -> addressed context`. The complete\nlookup and precedence rule is `docs\u002Freference\u002Fproduct-journey-sdlc.md#resolve-the-selected-path`\n(installed: `$OPENRIG_HOME\u002Freference\u002Fproduct-journey-sdlc.md#resolve-the-selected-path`).\nRead the selected addresses and source needed for this task; skills available in\nyour profile are capabilities, not a mandatory reading list. No composition means\nlight Part A. Role names and idle seats add no gates. Explicit rigor and authored\nwave boundaries retain their named checks.\n\nStartup files are address maps, not universal reading mandates. Do not preload\nunrelated planning\u002Freview doctrine or require per-file ACKs and quizzes. Skill\navailability is distinct from loading its full body. An old packet does not choose\nthe current work. Missing selected context is a named gap to resolve.\n\n## Proof standard\n\nIn a disposable fresh session, observe the actual reads and resulting next action.\nTest no-selection, explicit rigor, and wave-boundary cases. An edit is not evidence\nthat an already-running seat adopted it; adoption needs an authorized refresh or\nnext-launch observation. Do not clear or re-prime a live seat merely to test prose.\n\n## Cross-runtime startup paths (5; do not collapse)\n\nWhen a seat's startup is artifact-backed mental-model rebuild (active-work\nreentry case, distinct from reusable Agent Starters \u002F priming packs):\n\n- See the **cross-runtime restore\u002Freentry packet standard v0**\n- Source-trust ranking applies: **`rig whoami` > target rigspec > bounded latest transcript > full transcript > touched-files > `restore-summary.json`**\n\n## Memory surfaces consumed at startup\n\nInventory the selected startup inputs: AGENTS\u002Frole\u002FCULTURE overlays, replay\ncontext, and any declared restore packet or starter. Record where each comes\nfrom, whether it is current, and why this task needs it. An available skill or\nold packet does not select the work.\n\nUse the task-scoped startup path above to resolve current authority. Permission\nto read an input is not permission to rewrite its source; writes follow the\nactive project\u002Frig policy and assignment. Load\n`skills\u002Fopenrig-operating-model\u002FSKILL.md` with `rig context get` when deciding\nwhere durable context belongs.\n\n## Startup files vs skills (the distinction)\n\nPer `agent-startup-guide.md` (product reference doc) and the team handbook:\n\n| Startup files | Skills |\n|---|---|\n| Rig-specific, role-specific identity | Reusable SOPs \u002F methodology \u002F knowledge |\n| Tell agent WHO it is, WHAT it's working on, HOW this team operates | Tell agent HOW to do something (transferable across rigs) |\n| Examples: `role.md`, `CULTURE.md`, `startup\u002Fcontext.md` | Examples: `openrig-user`, `test-driven-development`, `vault-user` |\n| Authored per-rig | Authored once, used everywhere |\n\nDon't put skill content in startup files. Don't put identity content\nin skills. The 7-layer additive startup model (agent \u002F profile \u002F rig \u002F\nculture \u002F pod \u002F member \u002F operator) handles the layering.\n\n## See also\n\n- `mission-slice-sop` — load when starting assigned mission\u002Fslice work; the light artifact and handoff procedure for SPEC.md, NOTES.md, PROGRESS.md and proof. It does not choose the SDLC for the task.\n- `forming-an-openrig-mental-model` skill — orientation for new agents\n- `session-compaction-and-restore` skill — restore-time startup ingestion\n- `agent-starters` skill — reusable starter manifests that compose startup context\n- `openrig-operating-model` skill — placement and authority of durable context\n- `openrig\u002Fdocs\u002Freference\u002Fagent-startup-guide.md` (product reference doc; not a skill) — the 7-layer additive startup model + delivery hints\n","# Agent Startup and Context Ingestion\n\nHow an agent becomes useful after launch: **AGENTS.md overlays, role\nfiles, skills, rig specs, workflow specs, startup checklists, refocus\nmessages**, and the current `rig context` retrieval and profile surface.\n\nIt is the current concrete startup path inside the broader\ncontext-engineering-and-retrieval primitive. Startup gets a seat into\nthe right initial shape; context engineering is the larger question of\nhow a seat gets the right context for the work it is doing right now.\n\n**Most coordination failures are not tool failures; they are context\nfailures.** Agents need to know their role, operating mode, coordination\nconvention, boundaries, and current product intent. If startup context\nis scattered or stale, agents execute the wrong thing very efficiently.\n\n## Use this when\n\n- Authoring a new agent's startup files (role \u002F culture \u002F startup-context)\n- Refreshing a seat that's been running on stale guidance\n- Auditing whether agents got the current operating mode (not just what's in old files)\n- Designing the orchestrator → next-agent context-transmission shape\n- Building a startup map for OpenRig-building rigs\n\n## Don't use this when\n\n- The agent is being created via Agent Starter — the starter's manifest carries startup context\n- The work is artifact-backed mental-model rebuild from a packet — that's `session-compaction-and-restore`\n- The intent is to ship reusable startup content as a skill\n\n## Failure modes (4)\n\n1. **A new agent starts from an old rig spec and misses the current operating mode.** Specs go stale; current state must be visible at startup, not just historical config.\n2. **Guidance is written to a file that future agents read, but current agents are never told.** File edits don't propagate to running sessions. Cultural rollout (broadcast + fleet-changes-feed) is needed alongside file edits.\n3. **A startup file becomes a dumping ground and loses the map-to-canonical-sources role.** Startup should point AT canonical sources; it shouldn't TRY to be one.\n4. **The orchestrator transmits implementation instructions without preserving product intent.** Instructions decay; intent travels.\n\n## Task-scoped startup\n\nRun `rig whoami --json`, then resolve `project.yaml -> mission.yaml -> active\nslice.yaml -> selected component or wave map -> addressed context`. The complete\nlookup and precedence rule is `docs\u002Freference\u002Fproduct-journey-sdlc.md#resolve-the-selected-path`\n(installed: `$OPENRIG_HOME\u002Freference\u002Fproduct-journey-sdlc.md#resolve-the-selected-path`).\nRead the selected addresses and source needed for this task; skills available in\nyour profile are capabilities, not a mandatory reading list. No composition means\nlight Part A. Role names and idle seats add no gates. Explicit rigor and authored\nwave boundaries retain their named checks.\n\nStartup files are address maps, not universal reading mandates. Do not preload\nunrelated planning\u002Freview doctrine or require per-file ACKs and quizzes. Skill\navailability is distinct from loading its full body. An old packet does not choose\nthe current work. Missing selected context is a named gap to resolve.\n\n## Proof standard\n\nIn a disposable fresh session, observe the actual reads and resulting next action.\nTest no-selection, explicit rigor, and wave-boundary cases. An edit is not evidence\nthat an already-running seat adopted it; adoption needs an authorized refresh or\nnext-launch observation. Do not clear or re-prime a live seat merely to test prose.\n\n## Cross-runtime startup paths (5; do not collapse)\n\nWhen a seat's startup is artifact-backed mental-model rebuild (active-work\nreentry case, distinct from reusable Agent Starters \u002F priming packs):\n\n- See the **cross-runtime restore\u002Freentry packet standard v0**\n- Source-trust ranking applies: **`rig whoami` > target rigspec > bounded latest transcript > full transcript > touched-files > `restore-summary.json`**\n\n## Memory surfaces consumed at startup\n\nInventory the selected startup inputs: AGENTS\u002Frole\u002FCULTURE overlays, replay\ncontext, and any declared restore packet or starter. Record where each comes\nfrom, whether it is current, and why this task needs it. An available skill or\nold packet does not select the work.\n\nUse the task-scoped startup path above to resolve current authority. Permission\nto read an input is not permission to rewrite its source; writes follow the\nactive project\u002Frig policy and assignment. Load\n`skills\u002Fopenrig-operating-model\u002FSKILL.md` with `rig context get` when deciding\nwhere durable context belongs.\n\n## Startup files vs skills (the distinction)\n\nPer `agent-startup-guide.md` (product reference doc) and the team handbook:\n\n| Startup files | Skills |\n|---|---|\n| Rig-specific, role-specific identity | Reusable SOPs \u002F methodology \u002F knowledge |\n| Tell agent WHO it is, WHAT it's working on, HOW this team operates | Tell agent HOW to do something (transferable across rigs) |\n| Examples: `role.md`, `CULTURE.md`, `startup\u002Fcontext.md` | Examples: `openrig-user`, `test-driven-development`, `vault-user` |\n| Authored per-rig | Authored once, used everywhere |\n\nDon't put skill content in startup files. Don't put identity content\nin skills. The 7-layer additive startup model (agent \u002F profile \u002F rig \u002F\nculture \u002F pod \u002F member \u002F operator) handles the layering.\n\n## See also\n\n- `mission-slice-sop` — load when starting assigned mission\u002Fslice work; the light artifact and handoff procedure for SPEC.md, NOTES.md, PROGRESS.md and proof. It does not choose the SDLC for the task.\n- `forming-an-openrig-mental-model` skill — orientation for new agents\n- `session-compaction-and-restore` skill — restore-time startup ingestion\n- `agent-starters` skill — reusable starter manifests that compose startup context\n- `openrig-operating-model` skill — placement and authority of durable context\n- `openrig\u002Fdocs\u002Freference\u002Fagent-startup-guide.md` (product reference doc; not a skill) — the 7-layer additive startup model + delivery hints\n",{"name":5,"metadata":24,"description":7},{"openrig":25,"cli_surfaces_referenced":33},{"stage":26,"sibling_skills":27},"factory-approved",[28,29,30,31,32],"claude-compaction-restore","session-compaction-and-restore","agent-starters","session-source-fork","seat-continuity-and-handover",[34,35],"context","whoami","main","60438a272b68d091eddb01aad6e399af4996d536","agents\u002Fshared\u002Fskills\u002Fcore\u002Fagent-startup-and-context-ingestion",4,"https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world","https:\u002F\u002Fgithub.com\u002Fmvschwarz","https:\u002F\u002Fopenrig.dev","Building OpenRig. Running rogue AI civilizations. Debugging their coordination problems.","2026-11-12 13:31:38",{"source":8,"skillName":5,"commands":46,"prompt":56,"zipAvailable":57},[47,50,53],{"tool":48,"command":49},"npx","npx skills add mvschwarz\u002Fopenrig-world --skill agent-startup-and-context-ingestion",{"tool":51,"command":52},"bunx","bunx skills add mvschwarz\u002Fopenrig-world --skill agent-startup-and-context-ingestion",{"tool":54,"command":55},"pnpm","pnpm dlx skills add mvschwarz\u002Fopenrig-world --skill agent-startup-and-context-ingestion","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Fagents\u002Fshared\u002Fskills\u002Fcore\u002Fagent-startup-and-context-ingestion\u002FSKILL.md",true,[],[60,67,74,80,86,93],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7DMSEDCRREMBJF57RQB59HC","notes","Use when taking notes.","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Ftools\u002Ftest\u002Ffixtures\u002Finstall-verify\u002Fbundle\u002Fagents\u002Fteam\u002Fskills\u002Fnotes\u002FSKILL.md",[],[],{"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":19},"01M7DMSBZP9WE8MC09JDXFCR2F","software-for-agents","Use when older OpenRig instructions name software-for-agents or describe agent-in-the-middle or studio software under the former skill name.","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Frigs\u002Fworkshop\u002Fagents\u002Fworkshop\u002Fskills\u002Fsoftware-for-agents\u002FSKILL.md",[],[],{"skillNo":75,"name":29,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7DMS55VHEAGPXMHZJ7FX94R","Use when reasoning about what survives compaction\u002Fcontext-loss\u002Frestart\u002Fseat-refresh, designing a high-fidelity restore packet, or distinguishing native runtime resume vs fork vs artifact-backed mental-model rebuild. Covers the 4 failure modes that prevent honest restore (compacted seat drops hot potato; restore packet preserves details but loses product intent; runtime resume mistaken for handover or fork; rebuilt seat starts with stale instructions).","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Frigs\u002Fworkshop\u002Fagents\u002Fworkshop\u002Fskills\u002Fsession-compaction-and-restore\u002FSKILL.md",[],[],{"skillNo":81,"name":32,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M7DMRYQK0JTV7MMVHFWZ0BW0","Use when replacing a seat's occupant (rebuild\u002Fhandover\u002Fswap), reasoning about stable-seat-identity vs fluid-occupant-identity, choosing an old-occupant disposition (retire\u002Fadvise\u002Fshadow), or recording provenance for an occupant change. Two independent outcomes (continuityOutcome + seatBindingOutcome) and the 5 failure modes that prevent silent dishonesty.","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Frigs\u002Fworkshop\u002Fagents\u002Fworkshop\u002Fskills\u002Fseat-continuity-and-handover\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M7DMRP6RRGCJZSH9CE3PC1MY","retiring-and-inheriting-a-seat","Use when you are a sitting agent near your context threshold (~85%) and a PLANNED seat transition is due — retire deliberately and hand your seat to a fresh successor primed from a packet, rather than let compaction degrade you. Covers the handover packet, the append-only lineage ledger (one row per tenure), physical-seat continuity, the do-not-over-inherit framing, the optional warm-handoff window, and wake-v0 (consulting a retired predecessor). NOT for unplanned compaction\u002Fcrash recovery (session-compaction-and-restore \u002F claude-compaction-restore) and NOT the seat-binding primitive mechanics (seat-continuity-and-handover).","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Frigs\u002Fworkshop\u002Fagents\u002Fworkshop\u002Fskills\u002Fretiring-and-inheriting-a-seat\u002FSKILL.md",[],[],{"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":19},"01M7DMRJGN6Z8TZDR7JNHXH6KX","refocusing","Use when a long-running agent may have lost the product outcome, when context was compacted, when work has crossed a major boundary, or when a fresh path-based trace is needed before the next consequential action. Not for fresh-session orientation, waking an idle seat, or checkpoints.","https:\u002F\u002Fgithub.com\u002Fmvschwarz\u002Fopenrig-world\u002Fblob\u002Fmain\u002Frigs\u002Fworkshop\u002Fagents\u002Fworkshop\u002Fskills\u002Frefocusing\u002FSKILL.md",[],[]]