worktrail-sdd-workflow

Internal SDD workflow executor invoked by /go only. Do not call directly. Executes classified routes A–J: feature planning, implementation (parallel orchestration, PR generation), defect repair, refactoring, spec changes, investigation, and workflow evolution.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/worktrail-sdd-workflow/SKILL.md

Quality

Content

63%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.

A dense, highly actionable conductor skill with real validation gates and recovery loops — the strongest aspects are its executable commands and failure-path handling. It is dragged down by inlined historical bug narratives that bloat the token budget and by a two-level cross-bundle reference web that complicates navigation.

Suggestions

Move the historical bug narratives (orphaned-parent-record observations, datalena PR #2228 anecdote, Dispatch Contract restatement in the by:<dispatch-id> cell) into a separate design-notes reference and keep only the operative rule in SKILL.md.

Enumerate the ten completion states (or link to their authoritative list) so "finish --status <one of the ten completion states>" is executable without external lookup.

Fix or explain the Phase 1 → Phase 6 numbering jump (state that Phases 2–5 live in the pipeline references), and consolidate the scattered ../worktrail-go/references/*.md pointers so the reference chain stays one level deep from this skill.

Dimension	Reasoning	Score
Conciseness	The body is mostly high-signal imperatives, but pads tokens with historical narrative that the executor does not need: the orphaned-parent-record bug story citing "observations 5/6" of dead-dispatch-backlog-investigation.md, the "Gap observed live before this was code-enforced: datalena PR #2228" anecdote, and the giant by:<dispatch-id> table cell restating the Dispatch Contract. Matches "mostly efficient but includes some unnecessary explanation or could be tightened."	3 / 5
Actionability	Concrete, copy-paste-ready commands (worktrail-run-record start/scope-review, worktrail-land-pr with full args and --json), an exit-code table with recovery re-run instructions, and precise positional-argument tables. Minor gaps keep it from 5: "finish --status <one of the ten completion states>" never enumerates the ten states in this file, and $GATES, $RISK_LEVEL, $TITLE are used without local definition.	4 / 5
Workflow Clarity	The pipeline is clearly sequenced ("intake → record start → execute route → validate → PR → merge gate → outcome") with explicit checkpoints (scope-completeness gate, pre-PR gate, refused exit code) and real feedback loops (code_defect → repair per ci-watch-loop case 3 → re-run; review_threads_blocking → act → re-run). Held below 5 by the unexplained Phase 1 → Phase 6 numbering jump and validation detail delegated to the external worktrail-land-pr script.	4 / 5
Progressive Disclosure	Detail is deferred with anchored, clearly signaled references to the verified local bundle file (references/pipeline-details.md#new-pipeline) and sibling-skill playbooks (../worktrail-go/references/routes.md §A–J), but the chain nests two levels deep (pipeline-details.md itself points into worktrail-go/references/subagent-prompts.md at nearly every step), several references cross into other bundles and repo docs that cannot be assumed present, and long historical rationale is inlined that belongs in a separate file.	3 / 5
	Total	14 / 20 Passed

Description

78%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.

A strong description: concrete capability list, natural trigger terms, and unambiguous niche boundaries enforced by the "invoked by /go only" framing. The main weakness is that the when-to-use is expressed only negatively, and the route enumeration is compressed, costing a little specificity and completeness.

Suggestions

Add a positive explicit trigger clause, e.g. "Use when dispatched by /go with a route:A-J argument" instead of relying solely on the negative "Do not call directly" phrasing.

Add one or two natural synonyms for the action list (e.g. "bug fix" alongside "defect repair", "build" alongside "implementation") to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ("SDD workflow executor") and eight concrete capability areas ("feature planning, implementation (parallel orchestration, PR generation), defect repair, refactoring, spec changes, investigation, and workflow evolution"), but compresses the full route set into "routes A–J" rather than enumerating all ten, leaving minor coverage gaps.	4 / 5
Completeness	The "what" is explicit and concrete, and a "when" is stated via "invoked by /go only", but it is phrased negatively ("Do not call directly") rather than as an explicit positive trigger clause (e.g. "use when dispatched by /go with a route:X argument"), so the when could be more explicit.	4 / 5
Trigger Term Quality	"defect repair", "refactoring", "spec changes", "investigation", "feature planning", and "PR generation" are natural terms a user routing SDD work would say, and "/go" names the entry command. A few common synonyms ("bug fix", "build", "resume") are missing, so it falls between "some relevant keywords" and comprehensive.	4 / 5
Distinctiveness Conflict Risk	"Internal SDD workflow executor invoked by /go only. Do not call directly" carves out a clear niche and actively suppresses mis-triggering; minimal conflict risk with any other skill.	5 / 5
	Total	17 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: behindthedash/worktrail
Path: skills/worktrail-sdd-workflow/SKILL.md
Commit: 9e05b5c

Reviewed: 23 hours ago

Table of Contents

Discovery Implementation Validation

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.