Claude Code 2.1.285's Read deny rules let 4 of 11 file-reading routes reach the model in a developer's test, among them grep -r and CLAUDE.md @imports. The docs limit the rules to named paths and point anyone needing a hard block to the OS sandbox.
Reality
- Evidence62
- Adoption
- Insufficient
- Hype gap+15
- Incentives30
- Confidence60
Relipa's AI development workflow wraps a coding agent in six stages and fixes the tests before any code is written. Most of its published detail covers the workspace plumbing that runs before a model is ever called.
Reality
- Evidence35
- Adoption8
- Hype gap+5
- Incentives
- Insufficient
- Confidence40
Rulestack found a skill file that named a deleted pnpm script for 24 days while 28 commit-time shape tests passed. A 26-line check against package.json exits 1 on any missing name, so a hook can block the commit that deletes a referenced script.
Reality
- Evidence55
- Adoption5
- Hype gap+10
- Incentives
- Insufficient
- Confidence55
Huntress ran seven Rails tasks past Claude Fable 5.1 and watched it copy hand-rolled code out of its own repository. Changing the harness, in three cheap steps, took API recall from 48 percent to 100.
Reality
- Evidence58
- Adoption25
- Hype gap+15
- Incentives40
- Confidence55
Claude Code started reading AGENTS.md on September 18, but only where no CLAUDE.md sits in the working directory or any parent above it. Monorepos need a stub file per package or a change to the instructionFiles setting.
Reality
- Evidence55
- Adoption40
- Hype gap+10
- Incentives35
- Confidence55
A codeword test on Claude Code v2.1.273 put six same-sized memory files in six places. Three arrived at launch, two waited for a Read below them, and one needed an environment variable to arrive at all.
Reality
- Evidence64
- Adoption
- Insufficient
- Hype gap−12
- Incentives34
- Confidence57
Version 2.1.277 reads a shared AGENTS.md when no CLAUDE.md is in scope. The documented exceptions cover first sessions, telemetry-disabled runs, Bedrock, Vertex and Foundry, and the hooks governance tooling watches.
Reality
- Evidence68
- Adoption58
- Hype gap+12
- Incentives62
- Confidence64
A dev.to template for CLAUDE.md and AGENTS.md swaps style adjectives for Avoid/Preferred code pairs that name the exact helper an agent should call, on the argument that a code shape is the only part a model can imitate.
Reality
- Evidence34
- Adoption
- Insufficient
- Hype gap+22
- Incentives20
- Confidence50
A 29-session test on Claude Code v2.1.273 ran the same protected-directory rule two ways, as prose in CLAUDE.md and as a PreToolUse hook. Both held on a plain task. The comparison that separates them rests on four runs.
Reality
- Evidence58
- Adoption
- Insufficient
- Hype gap+10
- Incentives
- Insufficient
- Confidence55
Anthropic shipped the AGENTS.md fallback in Claude Code 2.1.277 as a flag-gated built-in mod. Two machines on the same pinned version can load different project instructions. The same week's digest also logs a reverted deny-rule fix.
Perspective Coverage
4 publishers
- Builder
- Builder 52%
- Operator
- Operator 39%
- Investor
- Investor 9%
Reality
- Evidence78
- Adoption62
- Hype gap−6
- Incentives58
- Confidence74
Claude Code's configuration-debugging page says nested memory files load on demand through the Read tool, so a report that the agent ignored the rules starts with /context and the question of whether the file was in the window at all.
Publishers:code.claude.com
Reality
- Evidence66
- Adoption
- Insufficient
- Hype gap+5
- Incentives55
- Confidence72
Anthropic's memory documentation sets a 200-line target per instruction file and reserves real blocking for a hook. Everything written in markdown arrives as context the model weighs against your last message.
Publishers:code.claude.com
Reality
- Evidence68
- Adoption
- Insufficient
- Hype gap+12
- Incentives60
- Confidence62
shinpr has taken claude-code-workflows through 133 releases, and the recent ones delete structure the models no longer need. His session reader then found three mandatory steps in his own repository that never ran.
Reality
- Evidence42
- Adoption18
- Hype gap+12
- Incentives58
- Confidence46
Claude can now read AGENTS.md, the file other agent tools already look for. XDA-Developers reports developers called the change small and long overdue, and the Claude-specific syntax that forced a second file has not moved.
Reality
- Evidence42
- Adoption28
- Hype gap+45
- Incentives55
- Confidence45
Claude Code now supports AGENTS.md through a built-in plugin with four loading modes. The default treats the two filenames as alternatives, and the mode itself is set in user or managed settings outside the repo.
Reality
- Evidence72
- Adoption45
- Hype gap−10
- Incentives58
- Confidence66
Claude Code now writes its own record of a developer's corrections and preferences. Anthropic's documentation keeps CLAUDE.md alongside it, and the one practitioner account available says the file's remaining job is narrower.
Reality
- Evidence48
- Adoption20
- Hype gap+14
- Incentives38
- Confidence46
An XDA Developers writer says his Claude allowance stretched four times further once he stopped letting the agent orient itself inside a 500-file repository. The post does not include token counts, so read it as one practitioner's log.
Reality
- Evidence24
- Adoption12
- Hype gap+35
- Incentives45
- Confidence38
Newer models ship with advice to keep the guidance file under 200 lines. The analyst who wrote those 1,042 lines argues each one logs context the model lacked, and that the real debt is a mechanical rule left sitting in prose.
Reality
- Evidence42
- Adoption22
- Hype gap−12
- Incentives38
- Confidence55
Go stops at exported and package-wide, so a helper meant for one file can be called from every file in the package. mpyw's linter checks that boundary and reports the crossing at both the declaration and the call.
Reality
- Evidence55
- Adoption10
- Hype gap+5
- Incentives60
- Confidence58
A dev.to write-up cut a 194,492-character CLAUDE.md to 17,283 by changing what triggers each rule to load, sorting them by what breaks if one stays unloaded at the moment it matters. The harness that checks the result runs one headless Claude Code session per prompt.
Reality
- Evidence55
- Adoption12
- Hype gap+25
- Incentives30
- Confidence60
Earlier coverage
- Scoping a Claude rules file to src/api/** kept 3,330 tokens out of the first request
Build · September 13, 2026 · 1 publisher
- Sixteen idle skills inside a used plugin escape /skill-doctor's prune list
Build · September 12, 2026 · 1 publisher
- An agent abbreviated its own changelog to bring CLAUDE.md eight bytes under a 45KB cap
Build · September 11, 2026 · 1 publisher
- Claude Code's auto memory relearned the same Chrome fix in three separate repositories
Build · September 10, 2026 · 1 publisher
- A six-assertion test blocks any commit that grows CLAUDE.md past 45,000 bytes
Build · September 4, 2026 · 1 publisher
- Undocumented rejections come back as confident agent proposals three weeks later
Build · August 29, 2026 · 1 publisher
- Anthropic's playbook concentrates human attention at gates that only evaluate file paths
Build · August 29, 2026 · 1 publisher
- A merge queue made the reviewing agent stricter than the agents that wrote the code
Product · August 28, 2026 · 1 publisher
- OpenCode has no memory subsystem. It has three mechanisms that fail differently
Build · August 25, 2026 · 1 publisher
- A memory layer beat CLAUDE.md by 22.2 points, and 54 of 72 test pairs never moved
Build · August 24, 2026 · 1 publisher
- Bundle archaeology beats "make it faster": 2.1 MB to 890 KB, one change per measurement
Build · August 24, 2026 · 1 publisher
- Anthropic cut 80% of Claude Code's system prompt and the evals did not move
Invest · August 23, 2026 · 1 publisher
- Test the rule change, not the model: a cheap A/B harness for agent standards
Build · August 23, 2026 · 1 publisher
- The personal agent is a folder, not a model: four files and less memory than you thought
Product · August 22, 2026 · 1 publisher
- Coverage at 80% was a price on human attention, and CI is the wrong place to charge it
Build · August 22, 2026 · 1 publisher
- The AGENTS.md file is an audit of the documentation you never wrote for humans
Build · August 19, 2026 · 1 publisher
- Opus 5 absorbed your verify prompts. The reading is still on your desk.
Build · August 19, 2026 · 1 publisher
- The 2-token prompt that billed 39,966 tokens: what a CLI subprocess hides
Build · August 18, 2026 · 1 publisher
- A 548KB CLAUDE.md burned 150,000 tokens before work started. The fix was a commit gate
Build · August 17, 2026 · 1 publisher
- The linter that passed everyone who ignored it and warned everyone who complied
Build · August 17, 2026 · 1 publisher
- Deleting a 1,350-line CLAUDE.md broke two rules out of twenty
Build · August 16, 2026 · 1 publisher
- The Context Tax: Your Developers Are Doing Unpaid Platform Work Every Session
Build · August 15, 2026 · 1 publisher
- A "Done." is a claim about the world, not a sentence you can grade
Build · August 15, 2026 · 1 publisher