trellis-continue

Resume work on the current task. Loads the workflow Phase Index, figures out which phase/step to pick up at, then pulls the step-level detail via get_context.py --mode phase. Use when coming back to an in-progress task and you need to know what to do next.

Quality

80%

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 ./.agents/skills/trellis-continue/SKILL.md

Quality

Content

85%

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

A tight, highly actionable resume command: exact commands, an unambiguous status-based routing table, and clean one-level-deep pointer to the canonical workflow file. The main gap is the absence of explicit validation/feedback checkpoints for the destructive tail of the flow (commit, archive, task.py start), which leaves workflow_clarity below the top anchor.

Suggestions

Add an explicit validation checkpoint before state-changing steps — e.g. verify the check actually passed and re-run it if not before moving to spec update/commit, mirroring a validate -> fix -> retry loop.

Make the commit/archive branch's prerequisites explicit (e.g. confirm git is clean, confirm check passed) so the destructive operations cannot be reached on stale state.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence: it gives the exact commands and a compact status-based routing table without explaining what git, commits, or phases are. Every line earns its place and there is no padding or concept re-explanation.	3 / 3
Actionability	Each step ships an exact, copy-paste-ready command (e.g. 'python3 ./.trellis/scripts/get_context.py --mode phase --step <X.X> --platform codex') plus a concrete decision table mapping status+artifact combinations to specific step numbers, leaving no ambiguity about what to run.	3 / 3
Workflow Clarity	Steps are clearly sequenced and the routing table is explicit, but there are no validation checkpoints for a workflow that drives destructive/state-changing operations (task.py start, commits, archive): status routing and 'only run task.py start after user confirms' are present yet there is no validate-then-proceed loop for the commit/archive steps.	2 / 3
Progressive Disclosure	SKILL.md acts as a concise entry point that signals one-level-deep references to the canonical guidance ('.trellis/workflow.md' for full phase detail) with no nested/deep references, and no bundle directories exist to require further splitting; the overview is appropriately separated from the detailed workflow.	3 / 3
	Total	11 / 12 Passed

Description

75%

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, action-oriented description that clearly states both capability and trigger, with concrete named commands anchoring the 'what'. Trigger term coverage and distinctiveness are the weaker spots: the phrasing is process-focused rather than spanning the natural variations a user might say, and the resume framing could overlap with sibling workflow skills.

Suggestions

Broaden trigger terms to cover natural user phrasings like 'continue', 'pick up where I left off', or 'what's the next step' so the skill surfaces for more varied requests.

Tie the trigger more tightly to Trellis-specific signals (e.g. an in-progress task in .trellis/) to reduce overlap with generic resume/task-management skills.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions: 'Loads the workflow Phase Index, figures out which phase/step to pick up at, then pulls the step-level detail via get_context.py --mode phase' — a clear sequence of specific operations rather than vague language.	3 / 3
Completeness	It explicitly answers both what (loads the Phase Index, finds the phase/step, pulls step detail via the get_context.py command) and when ('Use when coming back to an in-progress task and you need to know what to do next'), with an explicit 'Use when...' clause.	3 / 3
Trigger Term Quality	Triggers like 'Resume work', 'coming back to an in-progress task', and 'need to know what to do next' are natural, but it leans on process phrasing rather than the varied terms a user might actually say ('continue', 'pick up where I left off', 'what's next'), so coverage is good but not comprehensive.	2 / 3
Distinctiveness Conflict Risk	The task-resume niche is reasonably distinct, but 'Resume work on the current task' is generic enough that it could overlap with other task-management or workflow skills; the Trellis-specific artifacts (get_context.py, Phase Index) narrow it but the trigger phrasing is broad.	2 / 3
	Total	10 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: mindfold-ai/Trellis
Path: .agents/skills/trellis-continue/SKILL.md
Commit: 12e279a

Reviewed: 5 days 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.