context-anchored-fallback-report

Generate documents with write_file when retrieval tools fail, with explicit guardrails against task context drift

Quality

51%

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 ./benchmarks/gdpval/skills/write-file-fallback-report-enhanced-4bc5c2/SKILL.md

Quality

Content

56%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 delivers a clear, well-validated workflow with concrete, actionable patterns, but is significantly over-long and padded with redundant restatements of the same guardrail and basic markdown tutorial content, with no progressive disclosure to offset the monolith.

Suggestions

Collapse the redundant restatements: keep one authoritative statement of the CONTEXT_ANCHOR / 2-failure rule and remove its repetition across the warning, decision tree, Step 0/1, code example, guardrail tables, pitfalls, success criteria, and quick-reference card.

Remove the basic markdown structure tutorial in Step 3 ('Clear headers (##, ###)', 'Bullet points and numbered lists') — Claude already knows how to structure a markdown document.

Consolidate the overlapping Red Flags, Best Practices, Common Pitfalls, and Success Criteria tables into a single concise checklist to improve both conciseness and progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The 318-line body repeats the CONTEXT_ANCHOR / '2+ failures' rule across roughly ten sections and explains basic markdown ('Clear headers (##, ###)', 'Bullet points') that Claude already knows. Not 1 because genuine guardrail signal is present; not 3 because the redundancy and padding are pervasive rather than occasional.	2 / 5
Actionability	Provides concrete write_file/list_dir usage, a full Python example, a usable markdown template, and verification checklists. Not 5 because of placeholders ([CONTEXT_ANCHOR], [output_path]) and illustrative tool calls; not 3 because the guidance is concrete and largely usable rather than pseudocode.	4 / 5
Workflow Clarity	A clear Step 0→5 sequence with validation (list_dir), an integrity checklist, a feedback loop ('If any check fails, revise'), and success criteria. Not 5 because list_dir only confirms file existence rather than content/alignment quality; not 3 because checkpoints and a feedback loop are explicitly present.	4 / 5
Progressive Disclosure	No bundle files exist and the skill is a single 318-line monolith with redundant sections (Quick Reference Card, full code example, multiple overlapping tables) that could be consolidated. Not 4 because there is no file-level structure or progressive disclosure mechanism; not 2 because section headers provide reasonable internal organization.	3 / 5
	Total	13 / 20 Passed

Description

46%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 states a clear, specific action and a distinctive niche, but relies on internal/technical jargon rather than natural user trigger phrases and lacks an explicit 'Use when…' clause, capping its completeness and trigger-term quality.

Suggestions

Add an explicit 'Use when…' clause naming natural user situations, e.g. 'Use when web search or webpage access repeatedly fails during report generation and the user still needs a deliverable.'

Replace internal jargon ('retrieval tools fail', 'context drift') with user-facing phrasing and synonyms users would actually say ('web search is down', 'can't fetch the page', 'write the report anyway').

List one or two more concrete actions (e.g. 'Write structured markdown reports and verify the file was created') to lift specificity beyond a single action.

Dimension	Reasoning	Score
Specificity	Names one concrete, specific action ('Generate documents with write_file') plus a guardrail, but does not list multiple actions. Not 4 because there is no list of several specific actions; not 2 because the action names a specific tool rather than being generic.	3 / 5
Completeness	Has a clear 'what' ('Generate documents with write_file…') and a weak conditional 'when' ('when retrieval tools fail'), but no explicit 'Use when…' trigger clause, which caps completeness at 3 per the guidelines. Not 4 because the 'when' is an internal condition rather than explicit user-facing trigger guidance.	3 / 5
Trigger Term Quality	Terms like 'write_file', 'retrieval tools fail', and 'context drift' are internal/technical jargon; only the generic word 'documents' resembles a natural keyword. Not 1 because a recognizable output ('documents') and scenario are present; not 3 because no natural user phrasing or synonyms appear.	2 / 5
Distinctiveness Conflict Risk	The tool-failure-fallback trigger is a distinctive niche unlikely to wrongly fire, though the broad term 'documents' carries minor overlap with general document-generation skills. Not 5 because of that residual overlap risk; not 3 because the trigger condition is clearly specific.	4 / 5
	Total	12 / 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: HKUDS/OpenSpace
Path: benchmarks/gdpval/skills/write-file-fallback-report-enhanced-4bc5c2/SKILL.md
Commit: 3827781

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.