banthis
Banthis extends project memory with hard user prohibitions by writing a managed AGENTS.md/CLAUDE.md section and preserving all surrounding content. Use when the user says "stop doing X", "never do that again", "ban this", "remember not to", or "add a project rule".
What this skill does
# Banthis — durable behavioral bans Extend the local instruction surface with a named prohibition. The op-cell is **extend**: add one hard rule, keep existing project memory intact, and make the rule visible to future sessions. Banthis is for user frustration or correction that should persist. It is not a notes system, preference logger, or policy engine. --- ## When to Apply Apply immediately when the user signals a behavior should never recur: - “stop doing X” - “never do that again” - “ban this” - “remember not to …” - “add a project rule …” - repeated frustration with a specific agent behavior Certainty: - **HIGH** — explicit ban language or direct instruction to remember a prohibition. - **MEDIUM** — frustration plus a concrete repeated behavior. Ask only if the rule boundary is ambiguous. - **LOW** — stylistic preference, local task constraint, or one-off correction. Do not persist unless the user asks. --- ## When NOT Do not write a ban when: - the constraint applies only to the current command or current file; - the request is a normal implementation requirement; - the rule would be vague (“be better”, “stop being annoying”); - the rule duplicates an existing ban with no stronger wording; - the rule is unsafe, impossible, or conflicts with a higher-priority instruction. If a requested ban conflicts with an existing ban, surface the conflict instead of rewriting history. --- ## Workflow 1. **Extract the invariant.** Identify the forbidden behavior and the reason. No moralizing. 2. **Craft the title.** `<60` characters, noun phrase or imperative fragment, no markdown markers. - Good: `No silent scope shrink` - Bad: `User got mad because we keep doing dumb partial work` 3. **Craft the rule.** One line, direct prohibition, reason included: - `Do not X — reason.` 4. **Choose scope.** - Project-specific behavior: omit `--global`. - Agent-wide verbal tic or repeated generic habit: add `--global`. 5. **Write through the managed command.** From the skill directory, run: ```sh node scripts/banthis.mjs add "<title>" "Do not X — reason." ``` For user-wide bans: ```sh node scripts/banthis.mjs --global add "<title>" "Do not X — reason." ``` 6. **Report minimally.** Say which target file received the ban and whether it was added, updated, or unchanged. --- ## Target resolution The script resolves the write target deterministically: 1. `--global` always writes `~/.claude/CLAUDE.md`. 2. Otherwise, use the current working directory. 3. In the current directory, `AGENTS.md` takes precedence over `CLAUDE.md`. 4. If neither exists, create `CLAUDE.md`. No repository scan. No editor adapter. No external binary. --- ## Managed-section invariant The script owns only this range: ```md <!-- banthis:start --> ... <!-- banthis:end --> ``` Guardrails: - Never hand-edit inside the markers when the command can do it. - Never place user-supplied text containing banthis markers into a title or rule. - If the managed section exists, replace only that byte range. - If absent, insert after the first top-level `# H1`; if no H1 exists, prepend the section. - Preserve every byte outside the managed range except the minimal insertion boundary. The managed section contains a preamble asserting that bans win over the current request. Future sessions must surface conflicts instead of quietly violating bans. --- ## Command contract ```sh node scripts/banthis.mjs add <title> <rule> node scripts/banthis.mjs list node scripts/banthis.mjs remove <title> node scripts/banthis.mjs init ``` Shortcut: ```sh node scripts/banthis.mjs <title> <rule> ``` Flags: ```sh -g, --global target ~/.claude/CLAUDE.md -h, --help print usage ``` Results: - `added` — no case-insensitive title match existed. - `updated` — title matched case-insensitively and title or rule changed. - `unchanged` — exact title and rule already present. --- ## Anti-patterns - **Excess** — storing every preference as a hard ban. Bans are scarce and sharp. - **Graft** — adding a second memory mechanism, JSON cache, slash-command shim, or editor adapter. - **Sprawl** — scanning parent directories or rewriting unrelated docs. - **Soft ban** — phrasing as “try not to” or “prefer not to”. Use `Do not … — reason.` - **Hidden conflict** — obeying the current request while violating a persisted prohibition. --- ## Validation Gates Before yielding after a ban operation: 1. The target file contains exactly one `<!-- banthis:start -->` and one `<!-- banthis:end -->` pair. 2. The new or updated title appears once, case-insensitive. 3. The rule is one line and starts with `Do not` unless the user supplied a stronger exact wording. 4. Content outside the managed markers is unchanged except for first insertion. 5. `list` reports the expected title and rule preview.
Related in Writing & Docs
jax-development
IncludedUse this skill when the user is writing, debugging, profiling, refactoring, reviewing, benchmarking, parallelising, exporting, or explaining JAX code, or when they mention JAX, jax.numpy, jit, grad, value_and_grad, vmap, scan, lax, random keys, pytrees, jax.Array, sharding, Mesh, PartitionSpec, NamedSharding, pmap, shard_map, Pallas, XLA, StableHLO, checkify, profiler, or the JAX repo. It helps turn NumPy or PyTorch-style code into pure functional JAX, fix tracer/control-flow/shape/PRNG bugs, remove recompiles and host-device syncs, choose transforms and sharding strategies, inspect jaxpr/lowering/IR, and benchmark compiled code correctly.
nature-article-writer
IncludedDrafts, rewrites, diagnostically critiques, and style-calibrates primary research manuscripts for Nature and Nature Portfolio journals. Use when the user wants a Nature-style title, summary paragraph or abstract, introduction, results, discussion, methods, figure legends, presubmission enquiry, cover letter, reviewer response, or when a scientific draft sounds generic, jargon-heavy, structurally weak, or AI-ish and needs precise, broad-reader-friendly prose without inventing data, analyses, or references. Best for primary research articles and letters rather than reviews or press releases unless explicitly adapting one.
deckrd
IncludedDocument-driven framework that derives requirements, specifications, implementation plans, and executable tasks from goals through structured AI dialogue. Use when user says "write requirements", "create spec", "plan implementation", "derive tasks", "structure this feature", "break down into tasks", or "document this module". Also use for reverse engineering existing code into docs (/deckrd rev). Do NOT use for direct code writing — use /deckrd-coder after tasks are generated. Do NOT use when the user only wants to run or fix existing code without planning.
clinical-decision-support
IncludedGenerate professional clinical decision support (CDS) documents for pharmaceutical and clinical research settings, including patient cohort analyses (biomarker-stratified with outcomes) and treatment recommendation reports (evidence-based guidelines with decision algorithms). Supports GRADE evidence grading, statistical analysis (hazard ratios, survival curves, waterfall plots), biomarker integration, and regulatory compliance. Outputs publication-ready LaTeX/PDF format optimized for drug development, clinical research, and evidence synthesis.
handling-sf-data
IncludedSalesforce data operations with 130-point scoring. Use this skill to create, update, delete, bulk import/export, generate test data, and clean up org records using sf CLI and anonymous Apex. TRIGGER when: user creates test data, performs bulk import/export, uses sf data CLI commands, needs data factory patterns for Apex tests, or needs to seed/clean records in a Salesforce org. DO NOT TRIGGER when: SOQL query writing only (use querying-soql), Apex test execution (use running-apex-tests), or metadata deployment (use deploying-metadata).
accelint-ac-to-playwright
IncludedConvert and validate acceptance criteria for Playwright test automation. Use when user asks to (1) review/evaluate/check if AC are ready for automation, (2) assess if AC can be converted as-is, (3) validate AC quality for Playwright, (4) turn AC into tests, (5) generate tests from acceptance criteria, (6) convert .md bullets or .feature Gherkin files to Playwright specs, (7) create test automation from requirements. Handles both bullet-style markdown and Gherkin syntax with JSON test plan generation and validation.