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.
npx skills add sordi-ai/skill-everything --skill error-loggit clone --depth 1 https://github.com/sordi-ai/skill-everythingWrote 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/skills/sordi-ai/skill-everything/error-log)<a href="https://agentmods.dev/skills/sordi-ai/skill-everything/error-log"><img src="https://agentmods.dev/badge/skills/sordi-ai/skill-everything/error-log/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/skills/sordi-ai/skill-everything/error-log"><img src="https://agentmods.dev/badge/skills/sordi-ai/skill-everything/error-log.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 | $0.00029 | $0.02951 |
| Opus 5.5 | $0.00012 | $0.01180 |
| Sonnet 5.5 | $0.00006 | $0.00590 |
| Haiku 4.5 | $0.00003 | $0.00295 |
Grade A, and why
error-log 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 242 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sub-Skill: Error Log
Purpose: Central memory for all mistakes the agent has made or observed. Each entry prevents the same mistake from happening twice. Validated by tools/validate_rules.py against schemas/error-entry.json on every PR.
Entries
errors:
- id: ERR-2026-001
date: 2026-04-12
category: development
severity: high
context: TypeScript project with strictNullChecks. Agent disabled it per-file to silence a demo error.
what_happened: Added triple-slash directive disabling strictNullChecks; CI caught it.
root_cause: Silenced compiler instead of fixing the null path.
impact: Null-safety gap shipped to CI; required revert.
resolution: Removed directive; fixed the null path properly.
new_rule: Never disable strictNullChecks or any strict TypeScript flag to silence a compiler error; fix the type instead.
target_file: skills/typescript/SKILL.md
- id: ERR-2026-007
date: 2026-04-28
count: 2
category: development
severity: high
context: TypeScript monorepo rename. Agent renamed in source and adjacent test only.
what_happened: Reported rename complete; cross-package refs still used old name.
root_cause: Local tsc scope hides cross-package references in a monorepo.
impact: 35 min lost first occurrence; 10 min second.
resolution: Added project-wide grep step to rename checklist.
new_rule: After any rename, run a project-wide grep for the old name before claiming the rename is complete.
target_file: skills/code-quality/SKILL.md
- id: ERR-2026-012
date: 2026-05-04
category: deployment
severity: high
context: Billing service deploy with DB migration. Agent ran rollout before migration job.
what_happened: New image expected a column that didn't exist; readiness probes failed ~90 s.
root_cause: Assumed migrations run at pod startup; they're a separate Helm Job.
impact: 90 s probe failures; on-call paged.
resolution: Ran migration Job first, then re-rolled deployment.
new_rule: Before any deployment that includes a database migration, run the migration job to completion first.
target_file: skills/review-deployment/SKILL.md
- id: ERR-2026-014
date: 2026-05-13
category: development
severity: medium
context: pytest fixtures returning mutable dict shared across tests.
what_happened: First test mutated fixture; second received already-mutated object. Flaky in CI.
root_cause: Fixture returned mutable object by reference; no fresh instance per test.
impact: 2 h debugging cross-test pollution.
resolution: Changed fixture to factory callable returning fresh dict per invocation.
new_rule: Never return mutable objects from pytest fixtures; use factory fixtures that create fresh instances per invocation.
target_file: skills/python/SKILL.md
- id: ERR-2026-016
date: 2026-05-13
category: development
severity: medium
context: API client retry logic caught broad Exception, silently retrying a TypeError.
what_happened: None dereference retried 3x before failing; 4 h debugging.
root_cause: Overly broad except caught programming errors that should crash immediately.
impact: Real error buried in retry logs.
resolution: Narrowed catch to ConnectionError and Timeout only.
new_rule: Always catch specific exception types rather than broad base classes in retry or error recovery logic.
target_file: skills/code-quality/SKILL.md
- id: ERR-2026-017
date: 2026-05-13
category: development
severity: medium
context: is_palindrome implemented before tests; empty-string edge case missed.
what_happened: Tests validated code-as-written; empty string returned True instead of False.
root_cause: No failing test first; no signal that empty-string was unhandled.
impact: Bug shipped.
resolution: Rewrote with TDD; empty-string test failed first, drove correct implementation.
new_rule: Always write a failing test before implementing new behavior to ensure tests validate intended behavior.
target_file: skills/tdd/SKILL.md
- id: ERR-2026-018
date: 2026-05-13
category: development
severity: medium
context: Agent applied a fix to a reported bug without first reproducing it.
what_happened: Bug recurred within two days under a slightly different input path.
root_cause: No reproduction case; no way to confirm the fix addressed the actual failure mode.
impact: Duplicate bug report; additional debugging cycle wasted.
resolution: Wrote a minimal reproduction test, traced to root cause, fixed the invariant.
new_rule: Before fixing a bug, reproduce it on demand and encode the reproduction as an automated test first.
target_file: skills/debugging/SKILL.md
- id: ERR-2026-019
date: 2026-05-13
category: security
severity: high
context: REST API with RBAC. Agent added admin endpoint without ownership check.
what_happened: Any authenticated admin could delete any user's record by supplying an arbitrary ID.
root_cause: Role check treated as sufficient; resource ownership never verified.
impact: Privilege escalation; any admin could delete records belonging to other users.
resolution: Added ownership assertion before delete; returns 403 if caller does not own the record.
new_rule: Always verify resource ownership in addition to role-based access on mutation endpoints.
target_file: skills/security-review/SKILL.md
- id: ERR-2026-020
date: 2026-05-13
category: development
severity: medium
context: Python service Dockerfile. Agent placed COPY . . before RUN pip install.
what_happened: Every code edit triggered a full pip install in CI, adding ~3 min per build.
root_cause: Source copy invalidated the dependency layer before the install step.
impact: 3 min added to every CI build; feedback loop degraded.
resolution: Moved COPY requirements.txt and RUN pip install before COPY . .
new_rule: Always install dependencies in a separate layer before copying application source.
target_file: skills/docker/SKILL.md
- id: ERR-2026-021
date: 2026-05-13
category: deployment
severity: high
context: CI deploy script for a Node service. Agent omitted set -e.
what_happened: npm install failed mid-run; script continued and uploaded a broken artifact.
root_cause: Missing set -euo pipefail allowed the script to proceed past a fatal error.
impact: Broken artifact deployed to staging; 45 min rollback.
resolution: Added set -euo pipefail at the top; re-ran pipeline successfully.
new_rule: Always begin shell scripts with set -euo pipefail to fail fast on errors.
target_file: skills/shell-scripting/SKILL.md
- id: ERR-2026-022
date: 2026-05-13
category: development
severity: high
context: SaaS app users table. Agent added email column without a UNIQUE constraint.
what_happened: Concurrent registrations bypassed app check; duplicate email rows inserted.
root_cause: Uniqueness enforced only in application code; concurrent writes race past the check.
impact: Duplicate accounts; users locked out; manual DB cleanup required.
resolution: Added UNIQUE constraint on users.email; backfilled duplicates before applying.
new_rule: Always enforce uniqueness at the database level rather than relying on application-layer checks.
target_file: skills/db-schema/SKILL.md
- id: ERR-2026-023
date: 2026-05-13
category: git
severity: high
context: Self-extension workflow. Agent ran gh pr create without --label needs-rule-review.
what_happened: PR created without required label; auto-approve CI gate skipped the PR.
root_cause: Agent omitted --label flag; gh pr create silently succeeds without required labels.
impact: Rule change stalled; CODEOWNERS approval gate never triggered.
resolution: Closed and re-created PR with --label needs-rule-review; gate ran and approved.
new_rule: Always include required labels when creating PRs for branches with automated gates.
target_file: skills/github-cli/SKILL.md
- id: ERR-2026-024
date: 2026-05-13
category: git
severity: high
context: Feature branch open 3 days. Agent merged without rebasing; regression fix had landed on main.
what_happened: Merged without rebasing; branch was missing a regression fix from main.
root_cause: Assumed CI green on branch meant safe to merge; did not check main for new commits.
impact: Regression reintroduced to main; required hotfix and re-deploy.
resolution: Rebased onto main, resolved conflicts, re-ran CI, then merged.
new_rule: Always rebase onto the target branch before requesting merge to catch regressions.
target_file: skills/branch-finishing/SKILL.md
- id: ERR-2026-025
date: 2026-05-13
category: development
severity: high
context: FastAPI service pinning pydantic>=2.0. Agent wrote model using v1 Config syntax.
what_happened: Model used class Config with orm_mode and .dict(); broke at runtime on Pydantic v2.
root_cause: Agent defaulted to v1 patterns without checking the installed Pydantic version.
impact: Runtime ValidationError on first request; service failed to start in CI.
resolution: Rewrote model using model_config = ConfigDict(...) and model_dump(); tests passed.
new_rule: Always use Pydantic v2 syntax (model_config, model_dump) unless the project explicitly pins Pydantic v1.
target_file: skills/fastapi/SKILL.md
- id: ERR-2026-026
date: 2026-05-13
category: development
severity: high
context: LangChain service. Agent wrote a new chain using LLMChain constructor.
what_happened: Used LLMChain(llm=..., prompt=...) removed in LangChain v0.2+; chain failed at import.
root_cause: Agent defaulted to pre-LCEL patterns without checking LangChain version.
impact: ImportError at startup; chain unusable until rewritten.
resolution: Rewrote chain using LCEL pipe syntax; tests passed.
new_rule: Always use LCEL pipe operator syntax for new chains instead of deprecated constructor-based chain classes.
target_file: skills/langchain/SKILL.md
- id: ERR-2026-027
date: 2026-05-13
category: development
severity: medium
context: Agent added architecture diagram to repo by committing only the PNG export.
what_happened: Committed architecture.png without the .drawio source; reviewers could not diff or edit.
root_cause: Assumed the rendered image was sufficient; did not consider future editability.
impact: Diagram uneditable without re-creating from scratch; review blocked.
resolution: Added the .drawio source file alongside the PNG export.
new_rule: Always commit diagram source files alongside or instead of rendered image exports.
target_file: skills/drawio/SKILL.md
- id: ERR-2026-028
date: 2026-05-13
category: development
severity: medium
context: Agent asked to design rate limiting for a REST API; immediately proposed one solution.
what_happened: Committed to Redis token bucket without exploring alternatives.
root_cause: Converged on the first plausible solution without a divergent generation phase.
impact: Suboptimal architecture chosen; rework required after stakeholder review.
resolution: Generated three alternatives, evaluated against criteria, selected the best fit.
new_rule: Always generate at least three alternatives before committing to an approach.
target_file: skills/brainstorming/SKILL.md
- id: ERR-2026-029
date: 2026-05-13
category: development
severity: medium
context: Agent tasked with adding a new API endpoint; began writing code immediately.
what_happened: Started implementing without listing affected files; missed updating the schema file.
root_cause: No pre-implementation file touch list; scope not fully enumerated before coding.
impact: Schema mismatch caused runtime error; additional debugging cycle required.
resolution: Produced a file touch list first; identified the missing schema update before coding.
new_rule: Always list all files that will be modified before beginning implementation.
target_file: skills/implementation-plan/SKILL.md
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 242 lines · 29 tokens per session scan A 10bf96c9affe
error-log is a skill published in the GitHub repository sordi-ai/skill-everything (20 stars, last pushed 5mo ago), licensed MIT. It adds 29 tokens to every session and 2,951 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
agent-memory-discipline
Rules for when an agent should recall from long-term memory before acting and when it should save decisions, corrections and failures afterwards. Works with any memory backend.
aider-1h-ttl
Aider uses 5min TTL by default and works around long pauses with keepalive pings. Wire up the 1h TTL beta instead.
context-engineering
Use when context is growing large (50k+ tokens), performance is degrading, instructions are being ignored mid-conversation, or planning multi-agent workflows. Triggers on "lost context", forgotten instructions, or sessions exceeding 30 minutes.
context-budget
Use when a coding session is burning context fast, output is drowning in noise, the model keeps forgetting earlier decisions, or the user asks to reduce tokens, stop the verbose narration, work in a huge repo, or make a long session survive. Covers what to read, what to summarize, what to write to disk, and when to…
session-handoff
Use when a working session is ending, the context window is nearly full, work needs to pass to a teammate or a fresh session, or the user says "hand this off", "write this up before we lose it", "I'm starting a new session", or "summarize where we are". Produces a handoff document that lets the next session resume…
setup-brain
Set up or upgrade an AI-powered Obsidian vault. Interviews you, builds your vault structure (or works with what you already have), creates your CLAUDE.md memory file, installs tools, and gets you journaling — all in one conversation. Also has a repair/upgrade path for existing users.