filesystem-context

This skill should be used when agent work needs file-backed context: durable scratchpads, tool-output offloading, just-in-time discovery, cross-agent handoff files, filesystem memory, or cleanup policies for context stored outside the prompt.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Advisory

Suggest reviewing before use

Fix and improve this skill with Tessl

tessl review fix ./skills/filesystem-context/SKILL.md

Quality

Content

65%

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

A thorough, actionable skill with concrete code and good conceptual framing, weakened by verbosity, repetition across sections, an unreferenced bundle script, and the absence of explicit validation-checkpoint workflows.

Suggestions

Collapse the redundant Guidelines and Detailed Topics sections, and trim the opening paragraphs that re-explain the context-window limit, to tighten token use.

Reference scripts/filesystem_context.py from the body (or move its functions into the inline examples) so the bundle is discoverable rather than orphaned.

Add an explicit validate->fix->retry checkpoint sequence for the destructive/batch patterns (scratch cleanup, self-modification) to raise workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is mostly actionable but padded in places — the opening restates that context windows are limited (a concept Claude knows) and the Guidelines/Gotchas sections repeat points already made in Detailed Topics; it earns the mostly-efficient-but-could-be-tightened anchor rather than the lean 3.	2 / 3
Actionability	Provides concrete, executable artifacts — a complete `handle_tool_output` Python function, a YAML plan schema, directory layouts, and real grep/read commands like `grep -A 5 "error" terminals/1.txt` — matching the fully-executable, copy-paste-ready anchor above the pseudocode-only 2.	3 / 3
Workflow Clarity	Each pattern lists its own steps but there is no unified multi-step workflow with explicit validation checkpoints; validation guidance (existence checks, self-modification guards) is scattered and implicit, fitting the steps-listed-but-checkpoints-implicit anchor and below the validate->fix->retry 3 anchor.	2 / 3
Progressive Disclosure	The one reference is well-signaled one-level deep with a "Read when" trigger, but the main body is a ~290-line monolith that could be split further and the bundled scripts/filesystem_context.py is never referenced from the body, fitting the some-structure-but-could-be-better-organized anchor rather than the clean split at 3.	2 / 3
	Total	9 / 12 Passed

Description

85%

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 clear, third-person description with explicit use-when guidance and a distinct file-backed-context niche. Its main weakness is jargon-heavy trigger terms that users would rarely voice naturally.

Suggestions

Add natural-language trigger phrases users would actually say (e.g., "when the context window is full", "when tool output is too large to keep in context") alongside the technical capability list.

Consider trimming internal-feeling jargon like "just-in-time discovery" in favor of plain phrasings such as "load context on demand".

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete capabilities — "durable scratchpads, tool-output offloading, just-in-time discovery, cross-agent handoff files, filesystem memory, or cleanup policies" — matching the multiple-specific-actions anchor rather than the single-domain anchor at 2.	3 / 3
Completeness	The clause "This skill should be used when agent work needs file-backed context:" gives an explicit when-trigger, and the colon-list states what the skill does, satisfying both the what-and-when-with-explicit-triggers anchor; it is above 2 where the when is only implied.	3 / 3
Trigger Term Quality	Terms like "tool-output offloading", "just-in-time discovery", and "cross-agent handoff files" are technical jargon a user would rarely say naturally; relevant keywords exist but common user phrasings ("save output to files", "context window full") are missing, so it sits at the some-keywords-but-missing-variations anchor rather than 3.	2 / 3
Distinctiveness Conflict Risk	"file-backed context... stored outside the prompt" carves a clear niche distinct from the adjacent memory/compression/topology skills named in the body, making false-triggering unlikely; not the generic 1 or the still-overlapping 2 anchor.	3 / 3
	Total	11 / 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: muratcankoylan/Agent-Skills-for-Context-Engineering
Commit: c578e85

Reviewed: 7 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.