agent-pr-manager

Agent skill for pr-manager - invoke with $agent-pr-manager

3.94x

Quality

29%

Does it follow best practices?

Impact

75%

3.94x

Average score across 3 eval scenarios

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-pr-manager/SKILL.md

Quality

Content

42%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 has a recognizable structure and names real tools, but it is padded with marketing-style bullets, a stray duplicate YAML block, and pseudocode examples with corrupted paths that are not copy-paste executable. Destructive merge/approve workflows lack explicit validation checkpoints, and nothing is offloaded to reference files.

Suggestions

Delete the duplicate YAML block (name/tools/hooks) at the top of the body and the buzzword-only sections (Capabilities, Best Practices, Integration, Error Handling); state the same points as concrete rules in one or two lines each.

Fix the corrupted path separators ('src$index.js', '>$dev$null', 'integration$claude-code-flow-ruv-swarm') and replace the '...' placeholder arguments with concrete, copy-paste-ready examples.

Add explicit validation checkpoints to the merge workflow: after `gh pr checks`, only run `gh pr merge` when checks pass, and state what to do on failure (e.g. re-run tests, report, do not force-merge).

Dimension	Reasoning	Score
Conciseness	The ~155-line body is padded with buzzword sections that teach nothing ('Intelligent Review Strategy', 'Works seamlessly with', 'No single point of failure'), plus an entire duplicate YAML config block (name/tools/hooks) inlined at the top that belongs in frontmatter, not the body. This matches anchor 2 ('several unnecessary explanations or padded sections'); it is not a 3 because a large fraction of the file adds no executable guidance.	2 / 5
Actionability	Real tool names and parameters are given (swarm_init, task_orchestrate, gh pr create/view/review), but the 'examples' are non-executable JavaScript-style pseudocode of MCP calls, contain corrupted paths ('src$index.js', '>$dev$null', 'integration$claude-code-flow-ruv-swarm'), and use literal '...' placeholders — anchor 3 ('pseudocode instead of executable code; missing key details'). It clears anchor 2 because concrete tool names, arguments, and gh CLI commands are present.	3 / 5
Workflow Clarity	A rough lifecycle sequence exists (create → review → test → merge) and the merge example checks PR status first, but there are no explicit validation checkpoints or feedback loops for destructive batch operations (auto-approving reviews, squashing merges) — the 'Error Handling' section only asserts behaviors ('automatic retry logic', 'intelligent resolution') without steps. Per the rubric's cap, batch/destructive workflows without validation cannot score above 3.	3 / 5
Progressive Disclosure	No bundle files exist (no references/, scripts/, or assets/), so all content is inline in one file under clear section headers — 'some structure but could be better organized'. The inlined YAML config block and tool inventory are misplaced inline content, keeping it below anchor 4.	3 / 5
	Total	11 / 20 Passed

Description

16%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 is meta-boilerplate about how to invoke the skill rather than a statement of what it does or when to use it. It provides no capabilities, no natural trigger terms, and no usage conditions, making it nearly useless for skill selection. The real description ('Comprehensive pull request management with swarm coordination...') exists only in a second YAML block inside the body, not in the frontmatter field that matters.

Suggestions

Replace the frontmatter description with the actual capability statement from the body, e.g. 'Manages pull requests end-to-end: creates PRs, runs multi-agent code reviews, executes tests, and merges when checks pass' — concrete verbs instead of 'Agent skill for pr-manager'.

Add an explicit 'Use when...' clause covering natural trigger phrases, e.g. 'Use when the user asks to create, review, approve, or merge a pull request, or to check PR status or resolve merge conflicts'.

Remove the invocation meta-language ('$agent-pr-manager', 'invoke with') from the description; it describes the artifact, not the skill's purpose, and contributes no trigger terms.

Dimension	Reasoning	Score
Specificity	The description 'Agent skill for pr-manager - invoke with $agent-pr-manager' names no concrete action whatsoever — it describes what the artifact is, not what it does (no mention of reviews, merges, testing, or any capability). This matches anchor 1 ('Entirely vague; no concrete actions; pure abstract language') and is below anchor 2, which at least requires a named domain with generic actions like 'Processes PDF files'.	1 / 5
Completeness	There is no 'when to use' clause at all, and the 'what' is only vaguely implied by the name 'pr-manager' — anchor 2 ('Has a vague what and no when'). It is not a 1 because the name does at least gesture at pull-request management, but it falls well short of anchor 3, which requires a clear statement of what the skill does.	2 / 5
Trigger Term Quality	The only terms present are technical jargon ('Agent skill for pr-manager', 'invoke with $agent-pr-manager'); no natural phrases a user would say (e.g. 'pull request', 'review my PR', 'merge'). 'pr-manager' is a skill identifier, not a natural keyword, matching anchor 1's 'only technical jargon'.	1 / 5
Distinctiveness Conflict Risk	The narrow name 'pr-manager' distinguishes it from unrelated skills, but the description contains no triggering capability description, so it could overlap with any other PR/review-related skill ('Somewhat specific but could still overlap with similar skills'). It lacks the concrete capability list of anchor 4+.	3 / 5
	Total	7 / 20 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: ruvnet/ruflo
Path: .agents/skills/agent-pr-manager/SKILL.md
Commit: b14c79e

Reviewed: 3 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.