blindspot-pass

blindspot-pass is a skill for Claude Code from jellydn/my-ai-tools. It costs 24 tokens per session (1,392 once invoked), scanned A, original, MIT.

A pre-implementation investigation that searches a codebase, its history, and its existing patterns for hidden problems. A codebase is the project's source code and related files.

In plain words
What is it for?
Use it before working in an unfamiliar area, connecting systems, or handling complex or high-stakes changes.
Why use it?
It can uncover assumptions, edge cases, old workarounds, and integration risks before they cause problems during implementation.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter. Also seen: mentions Codex; built for cline; mentions OpenCode.

Good fit Use it before working in an unfamiliar area, connecting systems, or handling complex or high-stakes changes.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jellydn/my-ai-tools/blindspot-pass
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add jellydn/my-ai-tools --skill blindspot-pass
Clone the repo
git clone --depth 1 https://github.com/jellydn/my-ai-tools

Made for: Claude Code.

Wrote 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.

agentmods badge for blindspot-pass

README.md
[![agentmods](https://agentmods.dev/badge/skills/jellydn/my-ai-tools/blindspot-pass/github.svg)](https://agentmods.dev/skills/jellydn/my-ai-tools/blindspot-pass)
Your own site
<a href="https://agentmods.dev/skills/jellydn/my-ai-tools/blindspot-pass"><img src="https://agentmods.dev/badge/skills/jellydn/my-ai-tools/blindspot-pass/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.

agentmods 80×15 button for blindspot-pass

Your own site · 80×15
<a href="https://agentmods.dev/skills/jellydn/my-ai-tools/blindspot-pass"><img src="https://agentmods.dev/badge/skills/jellydn/my-ai-tools/blindspot-pass.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 24 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,392 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00024 $0.01392
Opus 5 $0.00012 $0.00696
Sonnet 5 $0.00005 $0.00278
Haiku 4.5 $0.00002 $0.00139

Measured 12d ago against content hash fdd84c4c9ccc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

blindspot-pass 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 12d ago.

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.

configs/amp/plugins/my-ai-tools-skills/skills/blindspot-pass/SKILL.md · 200 lines

How it starts

The opening of the file, as written. The whole thing — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Blind Spot Pass

When to Use

Use this skill before starting implementation when:

  • Working on an unfamiliar part of the codebase
  • Integrating with systems you don't fully understand
  • The task has high stakes or complexity
  • You sense there might be hidden gotchas

What It Does

A blind spot pass helps identify unknown unknowns that could derail implementation or lead to poor architectural decisions.

How to Execute

Step 1: Define Scope

Clearly state what you're about to work on:

  • Feature or change description
  • Affected modules/systems
  • Current understanding level

Step 2: Search for Context

Look in multiple places for relevant history:

# Recent changes to related files
git log --oneline --all -20 -- path/to/module/

# Search commit messages for keywords
git log --all --grep="auth\|oauth\|provider" --oneline

# Find related discussions in code comments
rg -i "TODO|FIXME|HACK|XXX" path/to/module/

# Search for error handling patterns
rg "catch|throw|error" path/to/module/ -A 3

Step 3: Identify Gotchas

Document findings in categories:

Technical Gotchas:

  • Edge cases in existing code
  • Performance considerations
  • Error handling patterns
  • Testing requirements

Architectural Gotchas:

  • Existing abstractions to follow
  • Integration points
  • Data flow patterns
  • State management approaches

Business Gotchas:

  • Implicit requirements from past discussions
  • Feature flags or phased rollouts
  • Backward compatibility needs
  • Security or compliance requirements

Step 4: Surface Questions

Generate questions that would change the approach:

  • "I see error handling uses X pattern; should I follow it?"
  • "There's a feature flag system; does this need one?"
  • "Previous auth work had Y consideration; does that apply here?"

Step 5: Present Findings

Format as a structured report:

## Blind Spot Analysis: [Task Name]

### Context Sources Checked
- [x] Git history (last 3 months, 45 commits in auth/)
- [x] Code comments and TODOs (found 3 relevant)
- [x] Error handling patterns (uses Result type)
- [x] Test coverage (integration tests in /tests/auth/)

### Key Findings

**Architectural Patterns**:
- All auth providers extend BaseOAuthProvider
- Must implement 3 lifecycle hooks: init, authenticate, refresh
- Config validation happens at provider registration

**Known Gotchas**:
- GitHub OAuth requires app installation per org (not per user)
- Refresh tokens expire after 6 months of inactivity
- Rate limiting uses Redis; need to handle connection failures

**Testing Requirements**:
- Mock responses in tests/fixtures/oauth/
- Integration tests use test credentials from .env.test
- Must test token refresh flow explicitly

### Questions Before Proceeding

1. Should GitHub provider support org-level installation?
2. How should we handle rate limit failures? (retry? user error?)
3. Is there a preferred library for GitHub OAuth, or implement from scratch?

### Recommended Approach

[Brief recommendation based on findings]

Read the full file on GitHub · 200 lines

Changes

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.

  1. 12d ago First seen · 200 lines · 24 tokens per session scan A fdd84c4c9ccc

Subscribe to this mod's changes

blindspot-pass is a skill published in the GitHub repository jellydn/my-ai-tools (120 stars, last pushed today), licensed MIT. It adds 24 tokens to every session and 1,392 once invoked, about $0.0001 per session on Opus 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 2026-08-30.