autoresearch:predict

autoresearch:predict is a command for Claude Code from hi0001234d/coding-agent-mcp-tools. It costs 29 tokens per session (549 once invoked), scanned A, original, Apache-2.0.

A multi-perspective code analysis command that asks several simulated expert viewpoints to examine code and combine their predictions into findings.

In plain words
What is it for?
Use it for pre-change analysis, security-oriented review, incremental review of changed files, or CI/CD checks that fail when findings reach a chosen severity.
Why use it?
It helps identify possible problems before changes are made by examining the code from different angles and, optionally, passing findings into debugging, security, fixing, shipping, or scenario work.

Command for Claude Code

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.

agentmods
npx agentmods add commands/hi0001234d/coding-agent-mcp-tools/predict
Clone the repo
git clone --depth 1 https://github.com/hi0001234d/coding-agent-mcp-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 autoresearch:predict

README.md
[![agentmods](https://agentmods.dev/badge/commands/hi0001234d/coding-agent-mcp-tools/predict.svg)](https://agentmods.dev/commands/hi0001234d/coding-agent-mcp-tools/predict)
Your own site
<a href="https://agentmods.dev/commands/hi0001234d/coding-agent-mcp-tools/predict"><img src="https://agentmods.dev/badge/commands/hi0001234d/coding-agent-mcp-tools/predict.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 549 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00029 $0.00549
Opus 5 $0.00015 $0.00275
Sonnet 5 $0.00006 $0.00110
Haiku 4.5 $0.00003 $0.00055

Measured 3d ago against content hash 3e3cdc66d3ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

autoresearch:predict 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 3d 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.

.claude/commands/autoresearch/predict.md · 38 lines

What it actually says

EXECUTE IMMEDIATELY — do not deliberate, do not ask clarifying questions before reading the protocol.

Argument Parsing (do this FIRST)

Extract these from $ARGUMENTS — the user may provide extensive context alongside flags. Ignore prose and extract ONLY flags/config:

  • --scope <glob> or Scope: — file globs to analyze
  • --chain <targets> or Chain: — comma-separated: debug,security,fix,ship,scenario
  • --depth <level> or Depth: — shallow (3 personas, 1 round), standard (5, 2), deep (8, 3)
  • --personas N — number of personas (3-8)
  • --rounds N — debate rounds (1-3)
  • --adversarial — use red team personas
  • --budget <N> — max total findings across all personas (default: 40)
  • --fail-on <severity> — CI/CD gate
  • --incremental — reuse existing knowledge files, update only changed files
  • --goal <text> or Goal: — focus area for analysis
  • Iterations: or --iterations N — integer for bounded mode (CRITICAL: run exactly N iterations then stop)

If Iterations: N or --iterations N is found, set max_iterations = N. Track current_iteration starting at 0. After iteration N, print final summary and STOP.

All remaining text not matching flags is the goal/focus description.

Execution

  1. Read the predict workflow: .claude/skills/autoresearch/references/predict-workflow.md
  2. If scope or goal is missing — use AskUserQuestion with batched questions per predict-workflow.md
  3. Execute the 8-phase predict workflow
  4. If bounded: after each iteration, check current_iteration < max_iterations. If not, STOP and print summary.
  5. If --chain is set, hand off to each chained command sequentially

Stream all output live — never run in background.

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. 3d ago First seen · 38 lines · 29 tokens per session scan A 3e3cdc66d3ca

Subscribe to this mod's changes

autoresearch:predict is a command published in the GitHub repository hi0001234d/coding-agent-mcp-tools (11 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 29 tokens to every session and 549 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.