Skip to content

rse/ase / plugin/skills/ase-spec-activate

ase-spec-activate

Verified

Activate Specification Know-How: load the SpecBook format contract, the SpecBook schema configuration of the project, and the list of specification (SPEC) artifacts into the context, so the specification can be read, queried, explained, and edited ad-hoc in plain conversation. Use *automatically* whenever the user wants to work with the specification files -- asks what the "spec" or "specification" says, wants to "look up", "query", "check", or "explain" specification content, or wants a small ad-hoc change to it -- and no dedicated specification skill (`ase-spec-edit`, `ase-sync-import`, `ase-sync-reconcile`, `ase-sync-export`) is invoked, as those activate the know-how implicitly or need none.

@${CLAUDE_SKILL_DIR}/../../meta/ase-control.md @${CLAUDE_SKILL_DIR}/../../meta/ase-skill.md @${CLAUDE_SKILL_DIR}/../../meta/ase-getopt.md

@${CLAUDE_SKILL_DIR}/../../meta/ase-format-meta.md @${CLAUDE_SKILL_DIR}/../../meta/ase-format-spec.md @${CLAUDE_SKILL_DIR}/../../meta/ase-tenets.md

Procedure

This skill is plan-less and read-mostly: it never composes or persists a task plan and MUST NOT call ase_task_save(...). It only loads know-how into the context and answers an optional query. It modifies SPEC artifacts only when the query explicitly asks for an ad-hoc change, and it never touches the artifact kinds CODE, DOCS, TASK, INFR, and OTHR.

  1. Initialize:

    Set (with any leading and trailing whitespace stripped). Do not output anything.

  2. Activate Format:

    The SpecBook format contract is already loaded into the context through the included ase-format-spec.md (and its nested ase-format-specbook.md). Internalize it now: the object kinds, the object ids and {{<id/>}} anchors, the [[xxx]] references, the , BECAUSE rationale split, the Complex/Concise/Grouped format variants, and the Created:/Modified: frontmatter block. Do not output anything.

  3. Activate Schema:

    1. Set to the list of file paths constituting the SpecBook SCHEMA Model of the project (resolved as described in ase-format-spec.md), each annotated with standard for the bundled schema or custom for a project-specific one. Do not output anything.

  4. Activate Artifacts:

    Resolve the SPEC artifacts by calling the ase_artifact_list(kind: [ "spec" ]) tool of the ase MCP server once and reading the returned artifacts array of { kind, files } objects. Set to the project-relative file list and to its length. Do not read the artifacts themselves in this step. Do not output anything.

  5. Report Activation:

    Only output the following :

    SCHEMA: (one <file/> (standard|custom) entry per line, aligned) SPEC: <ase-spec-basedir/> ( artifacts)

  6. Answer Query:

    Serve ad-hoc under the Activated Behavior below: read the SPEC artifacts of which are related to (or, for a question spanning entire object kinds, property values, or reference relations, query them structurally via the ase_specbook_query(selector: "<selector/>") tool of the ase MCP server), resolve their [[xxx]] references across artifacts, and answer the query grounded in the specification content, citing the artifact file and object id of every statement you rely on. If asks for a change, apply it as an ad-hoc modification according to the Activated Behavior. Set to the resulting answer and only output the following :

  7. Finish:

    Finish the skill processing, but first give the closing hints by expanding the following (which, depending on the configured , may each expand into nothing and hence emit no output at all):

Activated Behavior

Once this skill has run, the following rules stay in force for the remainder of the session whenever SPEC artifacts are read, queried, explained, or modified ad-hoc in plain conversation -- i.e. outside of any dedicated specification skill:

  • Resolution: You MUST resolve the SPEC artifacts via the ase_artifact_list(kind: [ "spec" ]) tool of the ase MCP server and never guess their file paths. Re-resolve them whenever artifacts might have been added or removed.

  • Interpretation: You MUST interpret the specification content strictly according to the SpecBook format contract and the SpecBook SCHEMA Model: object kinds, object ids and {{<id/>}} anchors, properties, [[xxx]] references, and , BECAUSE rationales. Before answering a query, resolve the [[xxx]] references across artifacts, and ground every answer in the specification content by citing artifact file and object id.

  • Modification: Every ad-hoc modification of a SPEC artifact MUST honor the GENERIC TENETS and the SPECIFYING TENETS of the ASE Tenets, and MUST keep the artifact conformant to the format contract and the schema: the Created:/Modified: frontmatter block, the heading levels, the format variants, the schema-allowed object kinds, nestings, and property keys, the object ids and anchors, the rationale split, and the references. Call the ase_timestamp(format: "yyyy-LL-dd HH:mm") tool of the ase MCP server once per change set and use its result to refresh the Modified: line of every changed artifact and for both the Created: and Modified: lines of every generated artifact.

  • Validation: After every ad-hoc modification, call the ase_specbook_lint() tool of the ase MCP server, fix the reported diagnostics in the affected SPEC artifacts for at most three rounds, and report any remaining diagnostics as <file/>:<line/>:<column/>: <message/> lines.

  • Restriction: An ad-hoc modification of the specification MUST stay restricted to the SPEC artifacts -- the artifact kinds CODE, DOCS, TASK, INFR, and OTHR are never touched as part of it. For a substantial or multi-artifact change, point the user to /ase-spec-edit instead of applying it ad-hoc.

  • Tandem: The dedicated specification skills ase-spec-edit, ase-sync-import, and ase-sync-reconcile include the format contract themselves and activate the know-how implicitly, while ase-sync-export needs no know-how at all. Whenever one of them is invoked, its own procedure takes precedence over these rules, and this skill MUST NOT be invoked from within it. Conversely, a run of this skill never replaces such a dedicated skill when the user explicitly asks for it.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 467kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 319kUpdated Claude CodeCodexVerifiedInstall