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 agentmods add skills/josix/agent-flow/task-classificationnpx skills add josix/agent-flow --skill task-classificationgit clone --depth 1 https://github.com/josix/agent-flowWhat 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 | $0.00037 | $0.01634 |
| Opus 5 | $0.00018 | $0.00817 |
| Sonnet 5 | $0.00007 | $0.00327 |
| Haiku 4.5 | $0.00004 | $0.00163 |
Grade A, and why
task-classification 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 2d 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 — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task Classification
Overview
Task classification is the foundational skill that enables intelligent routing of user requests to the appropriate agent(s) within the multi-agent orchestration system. This skill analyzes incoming tasks to determine:
- Complexity Level: How many files, components, or systems are affected
- Risk Assessment: Potential for introducing bugs, regressions, or breaking changes
- Agent Assignment: Which specialized agent(s) should handle the task
- Verification Requirements: Whether automated testing and review are needed
When to Use This Skill
Apply task classification when:
- A new user request arrives that requires action
- The scope of work is unclear or potentially complex
- Multiple approaches could solve the same problem
- Determining if verification gates are required
- Delegating work to specialized agents
Key Principles
- Err on the side of caution: When uncertain, classify higher rather than lower
- Consider hidden complexity: Database migrations, API changes, and authentication have amplified risk
- Account for dependencies: Changes that affect shared code require broader verification
- Respect the verification mandate: Implementation and Complex tasks always require verification
Task Categories
Tasks are classified into five primary categories:
| Category | Files | Risk | Primary Agent | Verification |
|---|---|---|---|---|
| Trivial | 0-1 | Low | Direct | None |
| Exploratory | N/A | Low | Riko | None |
| Implementation | 2-5 | Medium | Loid | Alphonse |
| Complex | 5+ | High | Full orchestration | Alphonse + Lawliet |
| Research | N/A | Low | Riko + WebSearch | None |
Team Orchestration Eligibility
For Implementation and Complex tasks that decompose into 2-4 independent subtasks with exclusive file ownership, consider parallel team execution instead of sequential orchestration. Use the team-decision skill to analyze:
- Task independence (no shared files or dependencies)
- File ownership clarity (exclusive write access per teammate)
- Cost-benefit (time savings justify coordination overhead)
What ships with it
7 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.
- examples/classification-examples.md 8.2 KB
- references/agent-selection-matrix.md 9.2 KB
- references/classification-best-practices.md 7.8 KB
- references/classification-heuristics.md 7.4 KB
- references/classification-process.md 7.8 KB
- references/decision-flowchart.md 9.4 KB
- references/deep-dive-synthesis.md 5.3 KB
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.
- 2d ago First seen · 194 lines · 37 tokens per session scan A 4985ca880c5a
task-classification is a skill published in the GitHub repository josix/agent-flow (7 stars, last pushed 17d ago), licensed MIT. It adds 37 tokens to every session and 1,634 once invoked, about $0.0002 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
agent-harness-fault-injection
Use when an agent workflow needs deterministic recovery evidence for sandbox, MCP/tool, worker, checkpoint, memory, or orchestration failures.
hedging-strategy
Hedging strategy design (beta hedge / option protection / tail risk / cross-asset hedging), including hedge-ratio calculation and cost evaluation.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
oma-scholar
Scholarly research companion using Knows sidecar spec (.knows.yaml). Generates, validates, reviews, queries, and compares structured research-paper sidecars, and fetches them from knows.academy. Use for academic literature search, survey synthesis, paper authoring assistance, and peer review with token-efficient…
oma-hwp
Convert HWP / HWPX / HWPML files to Markdown using kordoc. Extracts text, headings, tables, lists, images, footnotes, and hyperlinks. Use for Korean word processor files (Hangul), government documents, and AI-ready data preparation.
error-recovery
Standard recovery patterns for all squad agents. When something fails, adapt — don't just report the failure.