agentsclimarketplace

Gabe plan

Skill khujta/gabe-suite/skills/gabe-plan

KDBP-aware planning with lifecycle management + tier decision per phase (MVP/enterprise/scale) and optional HTML review artifacts for complex decisions. Usage: /gabe-plan [goal] [--full-catalog] [--html-artifact|--no-html-artifact]From its SKILL.md

Install
npx -y skills add khujta/gabe-suite --skill gabe-plan

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.0 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

Gabe Plan — KDBP-aware planner

Gabe execution contract (E1–E7)

This skill runs under the suite execution contract — E1 EVIDENCE · E2 RUN-BEFORE-✅ · E3 NO SILENT DOWNGRADE · E4 REUSE FIRST · E5 STATE SYNC · E6 MISSING ANCHOR = STOP · E7 REPORT WHERE — floors, not ceilings; a skill's own gate may be stricter, never looser. Full text: ../gabe-docs/references/execution-contract.md (if that file is missing, E6 applies — STOP).

What this does

KDBP-aware planner. Same planning logic as /plan, but persists to .kdbp/PLAN.md with lifecycle management plus a per-phase tier decision (MVP / Enterprise / Scale) rendered as a trade-off matrix. Every PLAN.md write also writes the .kdbp/PLAN.json machine mirror (phases, cells, tier, per-phase proof: field) in the same turn — read by session hooks and deterministic tooling. For complex plans it also creates a self-contained HTML review artifact as the human-facing entrypoint, while .kdbp/PLAN.md stays canonical.

Procedure

  1. Treat any text after the invocation as $ARGUMENTS.
  2. Read references/plan-spec.md (in this skill directory) IN FULL before executing — it is the binding spec: subcommand dispatch, tier-decision flow, PLAN.md write format, DECISIONS/LEDGER writes, and the shared auto-tick helper. If it is missing, E6 applies — STOP.
  3. Summary of the spec's main flow:
    • Step 0 — subcommand dispatch on first token of $ARGUMENTS: check → Step CHK (structural compliance + retrofit); update → Step UPD (modify active plan in place); complete/defer/cancel/replace → Step 1 branches; anything else is treated as a goal.
    • Step 0 (validate KDBP) — require .kdbp/, ensure archive/ and PLAN.md exist.
    • Step 0.5 (preset dispatch) — parses --html-artifact/--no-html-artifact/--html-path; --preset=mockup-project emits the canonical 13-phase mockup template (Step 3.PRESET) instead of free-form planning, then still runs the per-phase tier decision.
    • Step 1 — if an active plan exists, offer complete/defer/cancel/continue/replace; continue stops here.
    • Step 2 — gather context from .kdbp/BEHAVIOR.md (maturity, domain, tech); ask for the goal if none given.
    • Step 3 — draft the phase list; user confirms.
    • Step 3.5 — tier decision per phase (MVP/Enterprise/Scale): assemble the trade-off matrix from templates/gabe/tier-sections/* (Core always renders unfiltered; --full-catalog skips the Layer-2 LLM dimension filter), render the decision prompt, user picks a tier, log to DECISIONS.md (including per-dim tier overrides and suppressed dimensions), store tier + overrides in PLAN.md.
    • Step 3.75 — decide whether to create the HTML review artifact (--html-artifact forces, --no-html-artifact disables, otherwise complexity heuristics decide).
    • Step 4 — write .kdbp/PLAN.md (Goal, Context, Phases table, Phase Details, Current Phase, Dependencies, Risks, Notes, Review Artifacts, Runtime Evidence Checkpoints); Step 4b — write the .kdbp/PLAN.json machine mirror in the same turn.
    • Step 5 — append one PLAN row to the LEDGER.md thin session index.
    • Step 6 — archive mechanics for complete/defer/cancel/replace.
    • Step 7 — show the result.
    • Step CHK (/gabe-plan check) — zero-LLM structural compliance check of the active plan against the current spec shape, with a retrofit offer (LLM only fires if retrofit needs content generation).
    • Shared auto-tick helper — used by /gabe-execute, /gabe-review, /gabe-commit, /gabe-push to tick the Phases table's Exec/Review/Commit/Push column; idempotent, never silent on mismatch, always prints an enumerated skip code on precondition failure.

Output contract (summary)

Write .kdbp/PLAN.md with the full section set (Goal/Context/Phases/Phase Details/Current Phase/Dependencies/Risks/Notes/Review Artifacts/Runtime Evidence Checkpoints), the .kdbp/PLAN.json machine mirror, and the corresponding LEDGER.md thin-index row in the same turn (E5). Tier decisions, per-dim overrides, and suppressed dimensions get a DECISIONS.md entry with a stated reason. Emit the output-only **Gabe-Lens block** — never written to PLAN.md/REVIEW.md/LEDGER.md/PENDING.md/commits/docs unless another command already owns that write. When an HTML review artifact is created or refreshed, report its path. The full output contract in the spec is binding.

End every run with a single deterministic NEXT: /gabe-execute line (the phase's Exec command) — the routing contract; no other suggestions.

What ships with it: 2 files

61.5 KB alongside SKILL.md, 1 of them executable

references/

scripts/

Keep looking

Skills are one crate of 326,764. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.