gaia-mcp: Instructions file for Claude Code

Agent instructions

Read README.md and DEPRECATION.md before changing this repository. This repository is deprecated in favor of the bundled Skill Heaven plugin, but its published packages remain installable; do not start new product work here or pretend the deprecated package is the current plugin surface.

Organization dogfooding and feedback

Prefer the in-house Skill Heaven Agent Plugin or its actual supported Skills API surface when a genuine task need makes it relevant and the current harness makes it available. Native skills or the normal workflow may be a fallback when the in-house surface is unavailable or lacks a relevant usable skill; fallback does not count as Skill Heaven validation. This preference never overrides an explicit user invocation or a repository-required project skill. There is no skill quota: do not summon gratuitously, invent registration commands, pretend a client supports an unavailable surface, silently install anything, or edit user-global configuration. /summon is one-session context, not a permanent installation; preserve zero posture and human intent. If this repository's existing summon package surface is relevant, report its actual behavior rather than treating it as the portable plugin.

When friction appears during relevant real work or human feedback, capture a minimal local note immediately or as soon as practical. The exclusion list for public reports applies to capture as well as output: do not record raw prompts, transcripts, session logs, credentials, access tokens, raw tokenized content, personal data, private code, private URLs or paths, customer identifiers, or confidential details. Never record credentials or access tokens in feedback notes or drafts. Redact or synthesize before any external search query or delegate handoff. Store minimal drafts outside repository worktrees in caller-controlled local scratch. If uncertain, keep a constrained local note and ask the owner; do not export it. At task end, review pending notes; process only new actionable friction. Keep expected versus observed behavior, repo/commit, harness/version, route/source when known, a safe reproduction, impact, and evidence separate from inference. Identify human-reported facts as such. Retrieval score, retrieval rank, materialization, and exit code are not behavioral success or task outcome. Do not alter existing selection policy, labels, evidence semantics, or curator gates.