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 mskadu/opencode-agent-skills --skill permission-managergit clone --depth 1 https://github.com/mskadu/opencode-agent-skillsWrote 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/mskadu/opencode-agent-skills/permission-manager)<a href="https://agentmods.dev/skills/mskadu/opencode-agent-skills/permission-manager"><img src="https://agentmods.dev/badge/skills/mskadu/opencode-agent-skills/permission-manager.svg" alt="Measured on agentmods" 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 | $0.00023 | $0.00375 |
| Opus 5 | $0.00012 | $0.00187 |
| Sonnet 5 | $0.00005 | $0.00075 |
| Haiku 4.5 | $0.00002 | $0.00038 |
Grade A, and why
permission-manager 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 6d 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.
What it actually says
What I do
- Review and summarize currently always-allowed commands
- Suggest safe read-only commands for auto-approval
- Add or remove commands from the allow list in opencode.json
- Configure skill-level permissions (allow/deny/ask) with wildcard patterns
- Audit permission configs for security and usability
When to use me
Use this when optimizing opencode's permission settings, reviewing allowed commands, or configuring skill access controls.
Workflow Steps
- Read current config: Load
~/.config/opencode/opencode.jsonor project-levelopencode.json - Summarize permissions: Identify currently allowed commands and skill permissions
- Suggest additions: Propose safe read-only commands for auto-allow (see recommended list below)
- Apply changes: Edit the config to add/remove permission entries
- Validate: Ensure JSON is valid after changes
Complements opencode's built-in allow/deny/ask permissions by auditing current config and recommending adjustments through conversation.
Key Rules
- Never allow commands that modify files, commit, push, or change system state
- Use wildcards appropriately (e.g.,
git status*not justgit status) - Confirm with user before modifying permission config
- Distinguish between bash command permissions and skill permissions
- Keep config organized: group related commands together
How to trigger me
Use the Task tool with the permission-manager subagent type:
/permissions
Or in natural language, ask opencode to "manage opencode permissions" or "review allowed commands".
What ships with it
1 file 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.
- 6d ago First seen · 45 lines · 23 tokens per session scan A be43a4372b67
permission-manager is a skill published in the GitHub repository mskadu/opencode-agent-skills (1 stars, last pushed 3mo ago), licensed MIT. It adds 23 tokens to every session and 375 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 2026-08-31.
Other skills, from other repositories
architecture-diagram
Build interactive, click-through architecture diagrams for software systems — a self-contained single HTML file with animated step-by-step flows, mode toggles (dev/prod, offline/online, v1/v2), dark/light theme, and a side panel with payload details, plus a companion markdown description. Use when the user wants to…
rigorous-review
Rigorous, evidence-backed code review for high-stakes changes. Trigger on: rigorous review, thorough review, deep review, review this PR properly, review before merge, high-stakes review, "is this actually safe to merge", regulated or compliance code review, batch-review these PRs. Every finding carries quoted…
alloy-structural-analysis
Formal model-checking layer over Allium specs using Alloy 6 + Electrod + nuXmv. Trigger on: "rigorous structural analysis", "model-check this spec", "verify this Allium spec formally", "check for structural divergence with Alloy", "run the Alloy loop", "does the code actually satisfy this spec", temporal/CTL/LTL…
daily-briefing
Daily orientation and chief-of-staff briefing. Pulls from Slack (saved items), Gmail (important/starred), Google Calendar (today's events), Jira CLI (configured projects), and Todoist CLI, validates every action item, synthesizes a prioritized briefing using three consequence-based tiers, then produces a decisive…
jira
Manage Jira issues, sprints, and epics with the jira CLI (ankitpokhrel/jira-cli). Trigger on: jira, jira-cli, ticket, issue key (e.g. PROJ-123), epic, sprint, transition/move ticket, comment on ticket, assign ticket, JQL, create bug/story, "what's on my board", "my open issues". Use when the user wants to find, read…
scala
Enforces required tool order for Scala/Java projects. Compile/test: use the bloop CLI directly (bloop compile, bloop test) — never sbt compile/sbt test. Search/navigate: scalex first, then ast-grep, then a semantic/LSP-aware tool if one is available, then grep/ripgrep as last resort — this order is unconditional, even…