Borrowing it
Nothing to install: this file belongs to BlackbirdWorks/gopherstack. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/BlackbirdWorks/gopherstack/main/CLAUDE.mdgit clone --depth 1 https://github.com/BlackbirdWorks/gopherstackWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/blackbirdworks/gopherstack/claude-md)<a href="https://agentmods.dev/instructions/blackbirdworks/gopherstack/claude-md"><img src="https://agentmods.dev/badge/instructions/blackbirdworks/gopherstack/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/blackbirdworks/gopherstack/claude-md"><img src="https://agentmods.dev/badge/instructions/blackbirdworks/gopherstack/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01781 | $0.01781 |
| Opus 5.5 | $0.00712 | $0.00712 |
| Sonnet 5.5 | $0.00356 | $0.00356 |
| Haiku 4.5 | $0.00178 | $0.00178 |
Grade A, and why
gopherstack CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Instructions for AI Agents
pkgs/ usage and locking
Reuse pkgs/ instead of reimplementing in a service (e.g. awstime.Epoch would have prevented the QuickSight/IoT epoch-timestamp bugs).
- Lock granularity follows invariant granularity. One coarse
lockmetrics.RWMutexper service backend guards all its maps; AWS ops are cross-map transactions andSnapshot()needs one consistent view. safemap.Maponly for genuinely isolated single-map state (token stores, caches, janitor-swept stores). Never for backend resource maps: per-map locks tear cross-map state and risk deadlock.- No raw
sync.Mutex/sync.RWMutexin services. - Per-map init/Reset/Snapshot/Restore boilerplate is a lifecycle problem, not a locking one (planned fix:
gopherstack-drp). Persistence JSON field names must be preserved exactly. - ARNs:
arn, never hand-format. IDs:awsid.New("vpc-"), notuuid.NewString()[:N]. - Timestamps:
awstime;awstime.Epoch()for epoch-seconds JSON numbers. - Pagination:
page.Page[T], not hand-rolled NextToken/cursor logic. - Case-insensitive identifiers (e.g. RDS DBInstanceIdentifier):
strs(Fold,Equal,ContainsFold), not per-servicestrings.ToLower/EqualFold. - Outbound HTTP:
httpclient(own cloned transport); never&http.Client{}without a Transport (guarded byTestNoOutboundHTTPOnDefaultTransport). - All logging goes through
pkgs/logger. Nil-safe pointer deref:ptrconv. Tags:tags. Shared errors:awserr.
AWS parity rules
- No stubs: no no-op handlers, hardcoded/empty returns, fabricated IDs, or
&StubOutput{}. Every routed SDK op must mutate/read real backend state, return AWS-accurate shapes and error codes, and persist (wire into the service'spersistence.gobackendSnapshot). If an op can't be implemented yet, say so; never ship a half-working stub. - Verify wire shapes against the real
aws-sdk-go-v2serializers/deserializers, not your handler's own output. Known bug classes:- wrong XML list wrappers (
<member>vs<Item>), wrong root elements, missing response-root nesting when the output has no httpPayload member - ISO8601 strings where the JSON protocol expects epoch-seconds numbers
- a hardcoded
xml:"StubResponse"XMLName overriding the runtime root - stub registrations overwriting real handlers (EC2 uses
registerStubOpsIfAbsent; real handlers must win) - missing
errCodeLookupentries surfacing not-found as 500 InternalFailure
- wrong XML list wrappers (
- Unit tests are not parity proof. SDK-driven
test/integration/*_parity_test.gocaught 8 client-breaking wire bugs that green unit tests missed. Runtask build-linuxbeforego test ./test/integration/.... - Grep-based stub hunting has false positives: an empty envelope returned AFTER real backend logic is correct (void-result ops). Read the backend method before flagging; conversely, a real-looking op may filter a never-populated map.
- De-stubbing: remove the stub registration, delete the orphaned stub handler, drop it from bare-call stub test manifests, run
golangci-lint. - PARITY.md front-matter
items_still_open:is the only authoritative open list (gopherstack-anjf); nevergaps:/deferred:or body text. A pass that fixes an item must remove it in the same commit. Dated body sections are history only (task parity-lintcatches stale claims).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 124 lines · 1,781 tokens per session scan A 26c449392424
gopherstack CLAUDE.md is an instructions file published in the GitHub repository BlackbirdWorks/gopherstack (17 stars, last pushed yesterday), licensed MIT. It adds 1,781 tokens to every session, about $0.0071 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-07.
gopherstack CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| gopherstack CLAUDE.mdBlackbirdWorks/gopherstack | Instructions file | 17 | yesterday | MIT | A | 1,781 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| vscode ai-customization.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 1,907 | GitHub Copilot |