CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gsd-autonomous

Run all remaining phases autonomously — discuss→plan→execute per phase

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./pi/package-skills/gsd-autonomous/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

68%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tight, operationally specific orchestration wrapper: it gives exact dispatch mechanics, error-handling rules, and flag semantics while correctly deferring the detailed workflow to a referenced file. Its main weaknesses are implicit rather than explicit validation checkpoints for a batch, multi-phase run, and references that point at unresolved template placeholders with no bundle files present.

Suggestions

State the validation checkpoints explicitly instead of only "Preserve all workflow gates" — e.g. what must pass after each phase (plan review, execution verification) before advancing to the next, since this skill batches all remaining phases autonomously.

Include one concrete example of a `gsd_dispatch` call (role + task) so the dispatch contract is copy-paste unambiguous.

Resolve or explain the `__GSD_PI_ENGINE_ROOT__` placeholder paths in `<execution_context>` so the referenced workflow files are verifiably present at install time.

DimensionReasoningScore

Conciseness

The body is short and dense with operationally necessary Pi-specific rules (dispatch fail-closed behavior, @-include handling, flag semantics) that Claude would not know, and it assumes competence elsewhere. Minor over-explanation could be trimmed — e.g. the convoluted plan_review_convergence gate sentence under `--converge` ("the explicit flag overrides the gate; the gate governs the standalone command") and the note explaining that `<arguments>` is data, not instructions. Not a 5 because those passages don't fully earn their tokens; not a 3 because almost everything is non-obvious operational detail.

4 / 5

Actionability

Concrete, executable guidance dominates: exact tool name (`gsd_dispatch`) with its `role`/`task` parameters, explicit failure handling (STOP, surface error, optional-vs-required behavior), specific commands (`gsd-tools query init.milestone-op`, `gsd-tools query roadmap.analyze`), and fully specified flags. Minor gaps keep it from a 5: no example `gsd_dispatch` invocation and the actual workflow steps are deferred to the referenced autonomous.md rather than shown.

4 / 5

Workflow Clarity

The sequence is present ("For each phase: discuss → plan → execute" followed by "milestone audit → complete → cleanup") and dispatch failures have explicit handling, but validation checkpoints are only gestured at ("Preserve all workflow gates") rather than stated. Since this autonomously batches all remaining phases — a batch operation — the missing explicit validation/verification steps cap workflow clarity at 3 per the rubric.

3 / 5

Progressive Disclosure

Good structure: the body is a lean overview with clearly labeled sections (objective, execution_context, context, process), and the `<execution_context>` block signals one-level-deep references (@__GSD_PI_ENGINE_ROOT__/gsd-core/workflows/autonomous.md, ui-brand.md) that carry the detailed workflow. However, those paths are unsubstituted template placeholders and no references/ bundle exists alongside the skill, so the referenced material cannot be verified — a minor organization gap that keeps this below the well-signaled, easily-navigable anchor.

4 / 5

Total

15

/

20

Passed

Description

50%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description communicates a concrete action pattern (discuss→plan→execute across remaining phases) but omits the domain (milestone/roadmap-based development), any 'when to use' trigger guidance, and the natural synonyms users would say. All four dimensions sit at the midpoint, uniformly limited by vagueness about what the phases belong to.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user wants to finish remaining milestone phases hands-off, run all phases, or complete the roadmap without step-by-step involvement."

Name the domain so the description stands alone: reference milestone phases, ROADMAP.md, or the GSD workflow rather than bare "phases".

Include natural synonyms and artifacts users would mention — "milestone", "roadmap", "finish the milestone", "hands-off" — to improve trigger term coverage and distinctiveness.

DimensionReasoningScore

Specificity

The description names concrete per-phase actions ("discuss→plan→execute per phase") and a clear operation ("Run all remaining phases autonomously"), but the domain is left implicit — "phases" of what (milestone, roadmap, project) is never stated. This matches the anchor for naming 1-2 concrete actions without comprehensive coverage, and is above the 'names the domain but generic actions' anchor because the discuss→plan→execute cycle is specific.

3 / 5

Completeness

The 'what' is reasonably clear (autonomously execute all remaining phases via discuss→plan→execute), but there is no 'Use when...' clause or equivalent trigger guidance anywhere, which caps completeness at 3 per the rubric guideline. It is not a 2 because the 'what' is concrete rather than vague.

3 / 5

Trigger Term Quality

Relevant keywords exist ("run", "phases", "autonomously", "plan", "execute") that a GSD user might plausibly say, but common variations and synonyms are missing — no "milestone", "roadmap", "finish", "hands-off", or file references like ROADMAP.md. This lands at 'some relevant keywords but missing common variations', not the good-coverage anchor.

3 / 5

Distinctiveness Conflict Risk

"Run all remaining phases autonomously" is somewhat specific within a phase-based workflow toolset (distinguishable from single-phase skills by "all remaining" + "autonomously"), but "phases" is generic outside that context and could overlap with other multi-phase/autonomous-run skills in the same suite. Somewhat specific with residual overlap risk.

3 / 5

Total

12

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

open-gsd/gsd-core

Path

pi/package-skills/gsd-autonomous/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.