> ## Content Index > Fetch the complete content index at: https://www.implicator.ai/llms.txt > Use this file to discover other available public pages before exploring further. # Anthropic Adds a Claude Code Audit for Skills That Waste Context - URL: https://www.implicator.ai/anthropic-claude-code-skill-doctor-context-audit/ - Published: 2026-09-08T03:07:02.000Z - Updated: 2026-09-08T03:07:02.000Z - Description: Anthropic spent the summer telling developers their Claude configuration was too long. On September 4 it shipped a command that finds the excess. /skill-doctor reports which loaded skills a session never used, and what each one costs in context on every turn. - Author: Marcus Schuler - Tags: Tools & Workflows Anthropic has added a Claude Code audit that identifies loaded skills consuming context without being used and points developers to settings that can disable them. [Released on September 4, 2026](https://code.claude.com/docs/en/changelog?ref=implicator.ai), `/skill-doctor` tracks whether each eligible skill was invoked and shows its context cost. Developers can use the report to disable listing entries that consume context on every turn without being invoked. What Changed - Claude Code 2.1.261, released September 4, 2026, added \`/skill-doctor\`, which reports whether each loaded skill was ever invoked in a session and what it costs in context. - Anthropic's documentation states that every skill in the skill listing is added to context on every turn, whether or not Claude uses it. The listing budget defaults to 1% of the model's context window. - The audit follows two July 2026 changes to \`/doctor\`: version 2.1.205 made it a full setup checkup on July 8, and version 2.1.206 added a check on July 9 that proposes trimming checked-in CLAUDE.md files. - Anthropic says it removed over 80% of Claude Code's system prompt for Opus 5 and Fable 5 with no measurable loss on its coding evaluations. It has not published those evaluations. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/). ## A charge on every turn “Every skill in the skill listing adds to your context on every turn, whether or not Claude ever uses it,” [Anthropic’s documentation](https://code.claude.com/docs/en/skills?ref=implicator.ai) says. The listing gives Claude the names and descriptions it uses to decide which procedures apply to a request. That means an unused entry still occupies room that could hold code, tool results or the conversation itself. The report covers session skills other than bundled and enterprise entries. It flags listed skills that have never been invoked, tells the user where to switch them off and identifies plugins that have not been used recently. It requires Claude Code version 2.1.252 or later. Remote Control sessions opened from a phone or browser cannot generate the report. Anthropic also caps the combined `description` and `when_to_use` text for each listing entry at 1,536 characters. The listing budget defaults to 1% of the model’s context window. Developers can raise it through `skillListingBudgetFraction`; the documentation’s example sets the allowance at 2%. FREE WEEKDAY MORNING BRIEFING Follow what Anthropic ships to developers. The Implicator Morning Briefing filters the AI news cycle to the stories worth your attention and explains their consequences. From San Francisco, every weekday at 4:45 a.m. Pacific, 7:45 a.m. Eastern. Email address Send me tomorrow’s briefing Check your inbox for the confirmation link. About five minutes. No hype. No spam. ## The earlier cleanup The new audit follows two changes to `/doctor` in July 2026\. [Version 2.1.205](https://code.claude.com/docs/en/commands?ref=implicator.ai) made it a full setup checkup on July 8, while version 2.1.206 added a check on July 9 that proposes trimming checked-in `CLAUDE.md` files by removing material Claude can infer from the codebase. The command can also deduplicate local copies, move always-loaded guidance into skills or nested files that load when needed, and ask for confirmation before changing anything. “We removed over 80% of Claude Code’s system prompt for models like Claude Opus 5 and Claude Fable 5 with no measurable loss on our coding evaluations,” Thariq Shihipar, a member of technical staff at Anthropic, [wrote on July 24, 2026](https://claude.com/blog/the-new-rules-of-context-engineering-for-claude-5-generation-models?ref=implicator.ai). He said the team had been overconstraining the product through its system prompt, `CLAUDE.md` files and skills. Internal transcripts showed instructions colliding within one request, including “leave documentation as appropriate” beside “DO NOT add comments.” Anthropic replaced a blanket rule that code should default to no comments with a direction to match the surrounding code’s comment density, naming and idiom. Shihipar also wrote that examples can constrain newer models to a particular “exploration space.” ## What usage cannot prove Anthropic has not published the coding evaluations behind its “no measurable loss” claim. No independent measurement has been published showing what happens to output quality when the system prompt is cut. Know someone who'd find this useful? [✉️ Email it to a friend in one click](mailto:?subject=A%20newsletter%20I%20think%20you%27d%20like&body=This%20is%20one%20of%20maybe%20three%20newsletters%20I%20actually%20read.%20The%20rest%20just%20pile%20up%2C%20unread%2C%20judging%20me.%0A%0AAnd%20yes%2C%20this%20email%20mostly%20wrote%20itself%2C%20which%20is%20a%20little%20on%20the%20nose%20for%20an%20AI%20newsletter.%20Doesn%27t%20make%20it%20wrong.%20implicator.ai%20is%20good.%0A%0ASubscribe%20free%3A%20https%3A%2F%2Fwww.implicator.ai%2Fsubscribe%2F%3Futm%5Fsource%3Dnewsletter%26utm%5Fmedium%3Dforward%26utm%5Fcampaign%3Demail%5Fforward), or they can [subscribe free here](https://www.implicator.ai/subscribe/?utm%5Fsource=newsletter&utm%5Fmedium=forward&utm%5Fcampaign=forward%5Fto%5Fcolleague). The tool records whether a skill ran during a session and how much listing space it consumed, but not whether the skill improved the result. A rarely invoked but valuable procedure is therefore indistinguishable from dead weight in the report. One commenter on a [small Ask HN thread](https://news.ycombinator.com/item?id=49080135&ref=implicator.ai) disputed the idea that existing configurations had become incompatible. The thread had 5 points and 6 comments when captured on September 7, 2026\. “we moved from opus 4 to opus 5 without changing anything and no specific conflicts,” gojkoa wrote. The commenter called Anthropic’s advice a recommendation to shorten configuration, “not a backwards compatibility issue.” ## Controls stay with the user The broader `/doctor` command remains available even when the `disableBundledSkills` setting turns off the other bundled skills. `/skill-doctor` provides the narrower usage report, while `/doctor` reviews the configuration itself. Neither command silently edits a developer’s files. `/doctor` presents its findings first and asks for confirmation before making a change. Frequently Asked Questions What does /skill-doctor actually report? It flags skills loaded in a Claude Code session that were never invoked, shows what each costs in context, and says where to switch them off. It also identifies plugins that have not been used recently. It covers session skills other than bundled and enterprise entries. Which version added it, and are there limits? Claude Code 2.1.261, released September 4, 2026\. The documentation says the command requires version 2.1.252 or later. Remote Control sessions opened from a phone or browser cannot generate the report. Why does an unused skill cost anything? Every skill in the skill listing is added to context on every turn, whether or not Claude uses it, because the listing is how Claude knows what is available. Each entry's combined description and when\_to\_use text is capped at 1,536 characters to hold that cost down. What does /doctor change about CLAUDE.md? Since version 2.1.206 it proposes trimming checked-in CLAUDE.md files by removing material Claude can infer from the codebase, deduplicates local copies, and moves always-loaded guidance into skills or nested files that load on demand. It reports findings first and asks before changing anything. Has anyone verified that cutting context does not hurt quality? Not publicly. Anthropic's claim of no measurable loss rests on internal coding evaluations it has not released, and no independent measurement has been published. /skill-doctor itself records whether a skill ran, not whether running it helped. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/). [Microsoft’s David Fowler Says ‘Typing Code Is Absolutely Over’Microsoft distinguished engineer David Fowler said, “Typing code is absolutely over,” arguing that manual entry is becoming a smaller part of software work. Microsoft’s existing Aspire workflow gives The Implicator![](https://www.implicator.ai/content/images/2026/09/20260906-065844-fowler_typing_over.webp)](https://www.implicator.ai/microsoft-fowler-typing-code-over/) [Repo Radar: 5 GitHub Projects Worth Your WeekFive projects climbed GitHub this week and none of them is a model. Each sits in the layer a team can own outright: the rules an agent reads before it types, the diagrams it hands back, the workspace The Implicator![](https://www.implicator.ai/content/images/2026/09/20260903-102335-repo-radar-16.webp)](https://www.implicator.ai/repo-radar-5-github-projects-worth-your-week-16/) [Why AI Coding Agents Need Context Compaction Before 100 PercentAnthropic engineer Thariq Shihipar published a session-management guide for Claude Code on April 15, pairing the assistant's 1 million-token context window with a decision table that lists which commaThe Implicator![](https://www.implicator.ai/content/images/2026/05/2026-05-18-11.48.34-context_budget@2x.webp)](https://www.implicator.ai/anthropic-openai-google-tell-developers-to-budget-ai-context-windows/)