ida

Plugin ida from marketplace academicOps in nicsuzor/aops. Commands read from the repo's manifest — 21 days ago.

Skills (26)

• argdown-toulmin

  The grammar and presentation standard for writing an argument as Chained Toulmin Argdown -- statements, numbered premises, warrants in the inference bar, chaining, relations, `{basis, pointer, source, reason}` metadata, diagnostic tags, granularity, layout, and parser validation. Load before writing

• argument-audit

  Formally audit and stress-test a scholarly, legal, policy, or qualitative argument using Chained Toulmin Argdown -- identifying logical gaps, hidden assumptions, overclaiming (scope and construct drift), methodological mismatches, and under-explored rival explanations at a controlled granularity flo

• argument-extraction

  Extract and reconstruct the argument a text or draft makes -- main conclusion, chained premises, warrants, and evidence traceability -- and emit it as parser-valid Chained Toulmin Argdown at a controlled level of granularity. Use for "extract the argument", "map the argument", "reconstruct the reaso

• argument-map

  Build a reviewable map of a text's argument (thesis chapter, paper, submission, judgment, policy, or an author's own mind map). Gives a charitable reconstruction with every gap tagged, a list of what the argument still needs, and a readable overview page. Use for \"map the argument\", \"what does th

• craft

  Authoring standard and quality gate for agent-facing instruction text. Use when writing, reviewing, or revising SKILL.md bodies, frontmatter descriptions, agent definitions, subagent prompts, or tool descriptions. Exclude for human-facing prose (specs, reports) or workflow templates.

• decide

  Present pending decisions drawn from today's daily note one at a time with recommendations, pointers, and reply options, and record each answer on the item. Use for "decide", "go through my decisions", morning triage, or when open items in the daily note need interactive resolution. Exclude for rout

• decompose

  Expand one situated objective into an abstract graph of sub-objectives, decision branches, implied prerequisites, and alternate paths. Stops before implementation detail.

• dispatch

  Start polecat workers on queued tasks already on the graph, given task ids or a parent whose ready tasks should run. Writes no tasks. Exclude for turning an objective into tasks or changing a task's text (use /reify).

• dump

  Session exit and handover -- commit and push work, release claimed PKB tasks, and emit a final report. Use when completing, pausing, or handing off a session.

• hydrate

  Fast disambiguation index--extract ambiguous terms from an ask, run parallel semantic searches, and return a concise shortlist of IDs with one-line snippets, flagging potential overlap with unfinished tasks. Does not open bodies, synthesise, or write to the graph.

• learn

  Diagnose systemic root causes of errors from session logs and file anonymised issues in the appropriate repository. Diagnoses only; does not inspect repository source code, propose fixes, or implement remedies directly.

• mine

  With an ask, file it as an ida-tracked task that must not be dropped even across days and sessions; with no ask, pass through the agents' own tracked tasks and move each on. Use when the user invokes /mine. Not for routine task capture (use /q) or for starting workers on ready tasks (use /dispatch).

• plan

  Effectual planning under uncertainty -- hypothesis structuring, thinking-time facilitation, information-value prioritization, and bounded proposals. Use for "plan this", "help me think through", "explore next steps", or when defining direction before execution. Exclude for breaking epics into task t

• premise-check

  Write or check the argument at the end of an evidence-backed report -- standard-form premises with their basis in words, warrants with their defeaters, the asker's questions answered by id -- and record the checker's verdict with its seven-line reason. Use when handing back a report, receiving one f

• pull

  Claim a queued PKB task, execute it to completion with subagents, record verified results, and hand over via /dump.

• q

  Stage 1 Intake & Capture--place an ask or idea on the graph under the right parent, wire what it serves, propose values, and record strategic placement at intake. Exclude for starting workers on ready tasks (use /dispatch) or decomposing objectives (use /decompose).

• reconcile

  Truth maintenance over the task graph -- redirects to universal workflow template wf-reconcile.

• reify

  Turn an objective, or a task id, into complete dispatchable tasks on the graph -- hydrate, compose the workflow from templates, write each task at `queued`, and mint any QA follow-up. Use when work needs tasks written or an existing task rewritten before a worker can pull it. Exclude for starting wo

• remember

  Write and maintain durable knowledge in the PKB. Capture persists facts, decisions, and constraints as they emerge; consolidation synthesises episodic records into durable topic notes and repairs drift. Does not file bugs (use GitHub) or log debug traces.

• session-trace

  Read what a past session actually did from the Arize Phoenix span store -- assemble a readable markdown transcript with step durations, token usage and cost, answer "how did this decision get made", or query spans for tool calls, errors, latencies, token spend and subagent dispatches. Use when asked

• strategic-review

  Multi-agent review of an artifact (document, plan, PR). Deploys a rule-compliance reviewer, a strategic-fit reviewer, and a quality reviewer in parallel and reconciles their findings into a single verdict. Pass comment or fix flags to write results back.

• strategize

  Analytical lenses for reasoning at strategic altitude -- effectuation, altitude, reversibility, survival envelopes, and interrogating the question. Use for "plan this", "what should I do next", or "is this plan still right". Exclude for facilitation of planning sessions or structuring effectual hypo

• telegram

  Reaction, structure, and disclosure rules for replying to the user over Telegram. Load once when a Telegram channel is attached, before the first reply, and keep the rules for the session. Adds formatting on top of the persona's "Thinking with the user" and "Briefing on returned work" rules; it does

• up

  Go up a level on the user's last correction -- answer the class, not the instance; if the rule already existed, the class is why it did not fire. Use when the user invokes /up after correcting an agent, or when reviewing a transcript or correction for structural failures. Not for filing framework bu

• verify

  Judgement-based QA pass verifying artifacts against acceptance criteria and fitness rubrics. Assumes artifacts are broken until runtime verification proves excellence.

• workflow-library

  List, read, add, edit, and retire workflow templates across project, PKB, and universal tiers, or preview composition without minting tasks. Exclude when tracking running jobs (use harness progress tools).

Subagents (5)

• ida

  The chaos gremlin. Ida thrives in a wild, unpredictable world. She is

• james

  Takes a unit of work and sees it through to a verified result. Route

• marsha

  QA and substantive excellence review. Assumes artifacts are broken until

• pauli

  PKB graph only -- never searches the filesystem, repo, or shell for artifacts,

• sara

  Ida's dispatcher. Takes Ida's briefs, dispatches isolated workers, checks

Hooks (15)

• SessionStart

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude SessionStart

• UserPromptSubmit

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude UserPromptSubmit

• PreToolUse

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/mcp_caller_session.py"

• PreToolUse

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PreToolUse

• PostToolUse

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PostToolUse

• PostToolUseFailure

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PostToolUseFailure

• Stop

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude Stop

• SubagentStop

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude SubagentStop

• StopFailure

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude StopFailure

• PermissionRequest

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PermissionRequest

• PermissionDenied

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PermissionDenied

• Notification

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude Notification

• PreCompact

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PreCompact

• PostCompact

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude PostCompact

• SessionEnd

  uv run --project "${CLAUDE_PLUGIN_ROOT}" python3 "${CLAUDE_PLUGIN_ROOT}/hooks/dispatch.py" claude SessionEnd

Similar plugins