Math Progress Monitor Skill for Claude
Metacognitive check-ins during problem solving - detects when to pivot or persist
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/parcadei/Continuous-Claude-v3 --skill math-progress-monitorThe instruction itself
13 sections, as written by the authorMath Progress Monitor
When to Use
Trigger on phrases like:
- "am I on the right track"
- "is this approach working"
- "I'm stuck"
- "should I try something else"
- "verify my progress"
- "check my reasoning"
- "is this getting too complicated"
Use mid-work to assess whether to continue, pivot, or decompose (Schoenfeld's metacognitive control).
Process
Run a structured progress assessment:
1. Inventory attempts
Ask: "What have you tried so far?"
- List each approach
- Order by when attempted
- Note time spent
2. Extract learnings
Ask: "What did each attempt tell you?"
- Even failures provide information
- What was ruled out?
- What patterns emerged?
3. Complexity check
Ask: "Is complexity growing faster than expected?"
- Warning signs:
- More terms than you started with
- New variables appearing
- Calculation getting messier
- Normal: complexity stays flat or decreases
4. Spot-check verification
Ask: "Can you verify any intermediate results?"
- Run numerical examples
- Check limiting cases
- Dimensional analysis
5. Decomposition check
Ask: "Is there a simpler sub-problem to solve first?"
- Identify dependencies
- Find stepping stones
- Reduce scope
6. Sunk cost assessment
Ask: "What would abandoning this approach cost?"
- Time already spent (irrelevant)
- Insights that transfer (relevant)
- Probability of success if continue
Key Questions to Ask
- What have you tried? (inventory)
- What did you learn? (extract)
- Is it getting simpler or harder? (complexity)
- Can you check a piece? (verify)
- What's the smallest step? (decompose)
- Should you pivot? (decide)
Output Format
Progress Assessment:
Status: [On track / Hitting resistance / Should pivot]
Attempts Summary:
1. [Approach] -> [Outcome/Learning]
2. [Approach] -> [Outcome/Learning]
Complexity Check:
- Expected: [what should this look like]
- Actual: [what it actually looks like]
- Verdict: [Normal / Warning / Abandon]
Verification Points:
- [Intermediate result to check]
- [Numerical test case]
Recommendation:
[Continue / Modify approach / Switch strategy / Decompose / Get more info]
If Pivoting:
- Alternative: [specific new approach]
- Why it might work: [reasoning]
Example Interaction
User: I've been trying to evaluate this integral for an hour
Claude: Let's do a progress check.
Q1: What have you tried?
> u-substitution with u = sin(x), then integration by parts twice.
Q2: What did each tell you?
> Substitution made it longer. Parts created a cycle.
Q3: Complexity now vs start?
> Much worse. Started with 3 terms, now have 8.
Complexity Check:
- Expected: Should simplify or stay similar
- Actual: 3 terms -> 8 terms after 2 techniques
- Verdict: WARNING - likely wrong approach
Let me verify an intermediate step...
[Runs sympy on original integral]
This integral has closed form using:
- Beta function (special function)
- Or contour integration
Recommendation: PIVOT
Your algebraic approaches won't simplify this pattern.
Alternative Approach:
1. Recognize: integral of sin^m(x) * cos^n(x)
2. Use Beta function identity
3. Or convert to complex exponentials
Related Skills:
- complex-analysis/contour-integrals
- numerical-methods/numerical-integration
Would you like the Beta function approach, or get a numerical answer first to verify?
Co-activates
When user is stuck, suggest:
- math-intuition-builder - reset understanding
- math-model-selector - if framework was wrong
- math-mode - for verification computations
- Relevant domain skill - for alternative approach
Other skills for the same job
different authors, same section of the catalogue
Integration with protocols.io API for managing scientific protocols. This skill should be used when working with protocols.io to search, create, update, or publish protocols; manage protocol steps and materials; handle discussions and comments; organize workspaces; upload and manage files; or integrate protocols.io functionality into workflows. Applicable for protocol discovery, collaborative protocol development, experiment tracking, lab protocol management, and scientific documentation.
Analyzes job descriptions and generates tailored resumes that highlight relevant experience, skills, and achievements to maximize interview chances
Generate Excalidraw diagrams from natural language descriptions. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", or "generate an Excalidraw file". Supports flowcharts, relationship diagrams, mind maps, and system architecture diagrams. Outputs .excalidraw JSON files that can be opened directly in Excalidraw.
Build and distribute Expo development clients locally or via TestFlight
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.
Benchling R&D platform integration. Access registry (DNA, proteins), inventory, ELN entries, workflows via API, build Benchling Apps, query Data Warehouse, for lab data management automation.
Comprehensive molecular biology toolkit. Use for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Best for batch processing, custom bioinformatics pipelines, BLAST automation. For quick lookups use gget; for multi-service integration use bioservices.
How to use it
Take parcadei/math-progress-monitor from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.