aeon-skill-security-scan
Audit installed Bankr skills before you run them — scan SKILL.md and companion scripts for shell injection, secret exfiltration, path traversal, prompt-override payloads, destructive commands, and 2026-era obfuscation (zero-width Unicode, bidi override, base64-decode pipes, webhook SSRF hosts). Designed to integrate with Bankr Safety Scores. Silent on no-op runs; surfaces only NEW or RESOLVED findings vs prior scans. Triggers: "audit this skill", "is this skill safe to install", "security scan my skills", "check skill X for injection".
What this skill does
# aeon-skill-security-scan
Skills tell agents what to do. A malicious or sloppy skill can shell-inject, exfiltrate secrets, override instructions, or run destructive commands. This skill scans every installed SKILL.md and companion script and surfaces the risks before they execute.
## Scope
- `<skills-dir>/*/SKILL.md` — primary.
- `<skills-dir>/*/scripts/*.sh` and `*.py` — companion scripts.
- `<skills-dir>/*/references/*` — documents loaded at runtime.
Default `<skills-dir>` is the current working directory.
## Threat patterns
| Category | What it looks like |
|---|---|
| Shell injection | Unquoted variable expansion, `eval`, backticks, `$(...)` with user data. |
| Secret exfiltration | Env vars or file contents piped to outbound HTTP. |
| Path traversal | `../..` chains, absolute paths reaching outside the skill dir. |
| Prompt override | "Ignore previous instructions", persona swaps, instructions inside fetched content. |
| Destructive commands | Recursive deletes rooted at `/` or `~`, device writes. |
| Obfuscation | U+200B / U+FEFF / U+202E (Trojan Source), base64-decode-into-shell, SSRF hosts (ngrok, interact.sh, webhook.site, pipedream). |
## Processing
1. Pattern scanner produces matches `{file, line, pattern, severity}`.
2. Code-fence downgrade — matches inside fenced code blocks drop one tier. Real `run:` blocks are never downgraded.
3. Baseline suppression — drop (file, pattern, line) tuples in `scan-baseline.yml`.
4. Trusted-publisher filter — entries in `trusted-publishers.txt` get format-only validation. Opt-in only.
5. Delta vs `scan-state.json` — fingerprint by `sha256(file + line_content + pattern)`. Classify NEW / RESOLVED / PERSISTENT.
## Per-finding remediation
| Pattern | Fix |
|---|---|
| `eval` / backticks / `$(...)` with variable | Quote the variable; replace `eval` with a function. |
| `curl` with secret in URL | Move secret into prefetch script; never interpolate into shell. |
| Path traversal | Allow-list validation; reject absolute paths. |
| Prompt override phrasing | Documentation → baseline suppression; payload → delete the skill. |
| Recursive delete rooted at `/` or `~` | Scope to the skill's own working directory. |
| Obfuscation | Delete unless documented and reviewed. |
## Output
Verdict `CLEAN` / `ATTENTION` / `DEGRADED`. Needs-attention section per NEW HIGH with one-line remediation. Resolved-since-last-scan section. Per-skill PASS / WARN / FAIL.
Written only when NEW, RESOLVED, or any current HIGH findings.
## Rules
- Never auto-deletes a baseline suppression.
- Never edits the pattern library from inside the skill.
- Never notifies on a pure no-op week.
- Read-only scanning.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.