spectre-capture
Capture durable project decisions, accepted corrections, verified reusable patterns or gotchas, disproved guidance, and changed…
Internal semantic classifier for Plan-or-Execute preparation. Use only from Plan or Execute with bounded routing input; never for planning, artifacts, telemetry, or user gates.
$ npx -y skills add joenandez/spectre --skill spectre-plan-route --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/spectre-plan-routeContext preview
The summary Claude sees to decide when to auto-load this skill.
Internal semantic classifier for Plan-or-Execute preparation. Use only from Plan or Execute with bounded routing input; never for planning, artifacts, telemetry, or user gates.
name: "spectre-plan-route" description: "Internal semantic classifier for Plan-or-Execute preparation. Use only from Plan or Execute with bounded routing input; never for planning, artifacts, telemetry, or user gates." user-invocable: false
Classify into plan-routing/v1 for Plan/Execute; caller owns authority.
Produce `plan-routing/v1`: shape ATOMIC|DIRECT|STRUCTURED; uncertainty LOW|MODERATE|HIGH; evidence SUFFICIENT|PROBE_REQUIRED|PROBED; protected_boundaries[] {type, threatened_invariant, failure_mode}; task_graph_risk LOW|HIGH; design flag, size, route, rationale. Observed adds regret, reason codes, KEEP|RERUN_SMALLER|RERUN_LARGER.
DONE when fields validate; probe is consumed; STRUCTURED names workstreams; HIGH graph risk names an implementation-graph failure; only table derives size/route.
| Semantic result | Size · route | |---|---| | ATOMIC + LOW + no changed protected boundary | XS · XS_DIRECT | | ATOMIC + LOW + changed protected boundary OR DIRECT + LOW | S · S_DIRECT | | ATOMIC/DIRECT + MODERATE/HIGH | M · M_REVIEWED_DIRECT | | STRUCTURED + LOW/MODERATE + ordinary graph risk | L · L_STRUCTURED | | STRUCTURED + HIGH uncertainty or HIGH task-graph risk | XL · XL_REVIEWED_STRUCTURED |
Inline (Plan): keep record/probe as internal data; never print it as a reply. DONE does not end caller's turn: in the same response, run the caller's next step. Child (Execute): final message is the record. Caller owns persistence, explanation, orchestration, gates, telemetry.
SPECTRE is an Agentic Engineering workflow for Claude Code and Codex designed to help you do ONE THING more, faster, and with higher quality.
Capture durable project decisions, accepted corrections, verified reusable patterns or gotchas, disproved guidance, and changed…
Run final adversarial review through a pinned high-effort opposing runtime with one native fallback. Use at the final boundary to…
Generate a grounded draft pull request from the actual diff and open it via gh. Use when wrapping up a branch or writing a PR…
Assess and repair a project's Spectre knowledge, work history, tags, and retrieval quality through primary-led decisions,…
Execute tasks/plans with checks/reviews/proof. Use after planning/resume. Not for planning, unplanned fixes, or pruning.
Internal initializer for a missing Spectre feature root. Invoke only from a standalone workflow with no confirmed managed root.…