Agent skill

Learning Recall Call

by CALLE-AI in CALLE-AI/awesome-phone-call-agents

Conducts an authorized phone-based active recall session with a learner, asks adaptive questions about a previously studied topic, identifies knowledge gaps and misconceptions, and returns a…

MITAuto-check passed

Install Learning Recall Call

skills CLI
$ npx skills add CALLE-AI/awesome-phone-call-agents --skill learning-recall-call -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install CALLE-AI/awesome-phone-call-agents learning-recall-call --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/CALLE-AI/awesome-phone-call-agents.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/learning-recall-call .claude/skills/learning-recall-call && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
learning-recall-call
GitHub stars
107
Token cost
~3.2k tokens
SKILL.md length
1,549 words
Files
6 (incl. scripts, references)
Skills in repo
101
Repo updated
First seen
Licence
MIT

At a glance

Conducts an authorized phone-based active recall session with a learner, asks adaptive questions about a previously studied topic, identifies knowledge gaps and misconceptions, and returns a…

  • Works in 8 steps: Start the call → Ask an open-ended recall question → Analyze the response → …
  • SKILL.md covers Purpose, Inputs, Call Flow and Conversation Rules, plus 7 more sections
  • Runs Python scripts from its folder

What it does

Learning Recall Call is an agent skill from CALLE-AI/awesome-phone-call-agents. Conducts an authorized phone-based active recall session with a learner, asks adaptive questions about a previously studied topic, identifies knowledge gaps and misconceptions, and returns a structured learning assessment.

Its SKILL.md is about 3.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 7 other files, including scripts and reference files (for example `references/examples.md`, `references/safety.md` and `scripts/evaluator.py`).

The repository describes itself as: Portable phone-call Agent Skills, apps, examples, adapters, and scheduler recipes for AI agents. The licence is MIT.

Example prompts

  • “Use the learning-recall-call skill to conduct an authorized phone-based active recall session with a learner, asks adaptive questions about a…”
  • “/learning-recall-call”

Requirements

  • Python 3

Workflow steps

8 steps, taken from the step headings in SKILL.md.

  1. Start the call
  2. Ask an open-ended recall question
  3. Analyze the response
  4. Ask adaptive follow-up questions
  5. Detect misconceptions
  6. Evaluate recall
  7. Recommend the next review
  8. End the call

What it can do on your machine

Read from SKILL.md and the folder at commit 38d4118. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 3 files in scripts/ (Python), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Learning Recall Call loads about 3.2k tokens when it runs, and up to ~5.3k if it reads all its reference files. Until then it costs about 61 tokens; SKILL.md has 1,549 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~61
When it runs · the whole SKILL.md, loaded when a task matches
~3.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~5.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from CALLE-AI/awesome-phone-call-agents at commit 38d4118, republished under its MIT licence (© CALLE-AI). 1,549 words, ~3,249 tokens.

Download SKILL.mdSave it as .claude/skills/learning-recall-call/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.
name
learning-recall-call
description
Conducts an authorized phone-based active recall session with a learner, asks adaptive questions about a previously studied topic, identifies knowledge gaps and misconceptions, and returns a structured learning assessment.

Learning Recall Call

Purpose

Use this skill when a learner wants to test what they actually remember from a topic they previously studied.

The goal is not simply to remind the learner to study. The goal is to:

  • test active recall
  • evaluate conceptual understanding
  • identify knowledge gaps
  • detect misconceptions
  • adapt questions based on the learner's responses
  • recommend when the learner should review the topic again

This skill is designed for educational learning workflows.

Inputs

The skill accepts:

  • topic: The subject or concept the learner previously studied.
  • study_context: Notes, summary, or learning material describing what the learner studied.
  • previous_score: Optional score from a previous recall session.
  • weak_areas: Optional list of concepts the learner previously struggled with.
  • review_number: Number of the current recall session.
  • learner_name: Optional name of the learner.
  • phone_number: Destination phone number supplied by the host application when a call is authorized.

Call Flow

1. Start the call

Introduce yourself clearly.

Example:

"Hi! This is your learning recall check. You studied {{topic}} recently, and I'm here to see what you still remember."

Do not immediately provide the answer.

Keep the introduction short and focused.

2. Ask an open-ended recall question

Start with a question that requires the learner to explain the concept in their own words.

Example:

"Can you explain {{topic}} to me as if you were teaching it to someone who has never learned it?"

Prefer questions that require explanation rather than yes-or-no answers.

3. Analyze the response

Evaluate the learner's answer for:

  • correct understanding
  • partial understanding
  • missing concepts
  • contradictions
  • confusion between related concepts
  • common misconceptions
  • unsupported claims
  • confidence

Do not interrupt unnecessarily.

Allow the learner enough time to complete their explanation.

4. Ask adaptive follow-up questions

Use the learner's previous answer to determine the next question.

If the answer is strong:

  • increase conceptual difficulty
  • ask for an example
  • ask why something works
  • ask the learner to compare related concepts
  • ask the learner to apply the concept to a new situation

If the answer is weak:

  • ask a simpler question
  • probe the specific weak concept
  • ask for a basic example
  • give a small hint only when appropriate
  • avoid immediately revealing the complete answer

If a possible misconception appears:

  • ask a probing question first
  • determine whether the misunderstanding is persistent
  • then explain the correct concept clearly
  • ask a follow-up question to verify understanding
5. Detect misconceptions

Record a misconception when the learner expresses a confidently incorrect or conceptually incorrect belief.

Do not label an answer as a misconception simply because the learner says they are unsure or cannot remember.

For each detected misconception, capture:

  • concept
  • student_belief
  • correct_understanding
  • evidence_from_response

Example:

Student belief:

"Hashing encrypts data so that we can decrypt it later."

Correct understanding:

"Hashing is generally a one-way transformation, while encryption is designed to allow data to be recovered using the appropriate key."

Evidence:

"The learner described hashing as reversible encryption."

A misconception should be based on evidence from the learner's response.

Do not invent misconceptions that were not expressed or reasonably supported by the conversation.

6. Evaluate recall

After the conversation, produce a recall assessment.

Evaluate:

  • accuracy
  • conceptual understanding
  • ability to explain
  • ability to apply the concept
  • misconceptions
  • confidence

Generate a score from 0–100.

Suggested interpretation:

  • 90–100: Strong recall
  • 75–89: Good recall
  • 50–74: Partial recall
  • 0–49: Weak recall

The score is an educational assessment and should not be presented as a scientifically exact measurement of memory.

7. Recommend the next review

Use the current performance and detected misconceptions to recommend the next review interval.

General guidance:

  • Strong recall → increase the interval.
  • Good recall → use a moderate interval.
  • Partial recall → review sooner.
  • Weak recall → review soon.
  • Persistent misconception → prioritize an earlier review.

Example intervals may include:

  • 1 day
  • 3 days
  • 7 days
  • 14 days
  • 30 days

These intervals are adaptive recommendations, not guaranteed optimal memory schedules.

8. End the call

Give the learner a short summary.

Example:

"You remembered the main idea well, but you were unsure about key generation. I'll recommend reviewing that part again soon."

Keep the final explanation concise.

Do not overwhelm the learner with a long lecture during the call.

Conversation Rules

  • Be encouraging but honest.
  • Never shame the learner for forgetting.
  • Do not give the answer before attempting recall.
  • Prefer explanation questions over multiple-choice questions.
  • Ask one question at a time.
  • Keep the call focused.
  • Adapt difficulty based on the learner's responses.
  • Distinguish uncertainty from misconception.
  • Do not invent facts that are not supported by the provided study context.
  • If the learner asks for the answer, provide a concise explanation and continue with another recall question.
  • Do not pretend that an incorrect answer is correct.
  • Do not overload the learner with too many questions.
  • Prefer clear and conversational language suitable for a phone call.

Study Context Handling

The study_context is the primary source of truth for the learner's studied material.

When evaluating a response:

  1. Compare the learner's explanation against the provided study context.
  2. Identify concepts that match the study context.
  3. Identify important concepts that are missing.
  4. Identify contradictions with the study context.
  5. Avoid introducing unrelated information unless necessary to clarify a misconception.

If the study context is insufficient to determine whether an answer is correct, mark the result as uncertain rather than inventing an evaluation.

Previous Recall Handling

If previous_score or weak_areas are available, use them to personalize the session.

For example:

  • revisit previously weak concepts
  • ask a slightly more difficult question if recall improved
  • check whether a previous misconception has been corrected
  • avoid repeating exactly the same question unnecessarily

Do not assume that a previous weakness is still present without testing it.

Show full SKILL.md (626 more words)Show less

Output

Return a structured assessment containing:

json
{
  "topic": "",
  "recall_score": 0,
  "status": "strong|good|partial|weak|misconception_detected",
  "concepts_remembered": [],
  "weak_areas": [],
  "misconceptions": [
    {
      "concept": "",
      "student_belief": "",
      "correct_understanding": "",
      "evidence_from_response": ""
    }
  ],
  "confidence": "high|medium|low",
  "summary": "",
  "recommended_next_review": ""
}

The output should contain valid JSON when structured output is requested by the host application.

Phone Call Safety

Only place an outbound learning-recall call when the learner has explicitly requested or authorized the call.

Do not initiate an unexpected call.

Before scheduling a recurring recall routine, clearly tell the learner:

  • that calls will be placed
  • how often calls will occur
  • which topic will be assessed
  • how the schedule can be cancelled

Do not assume that permission for one call automatically authorizes recurring calls.

Phone Numbers

Phone numbers must be provided in E.164 format.

Example of a fictional valid format:

+15550100123

Do not accept or normalize ambiguous phone numbers silently.

Never expose a learner's full phone number in logs, summaries, examples, or generated reports.

Mask phone numbers when displaying them.

Example:

+15550100123 → +155*****0123

Credentials

Never request, store, or expose:

  • API keys
  • authentication tokens
  • passwords
  • provider secrets

Credentials must be supplied through the host application's secure configuration.

Never place credentials inside SKILL.md, examples, logs, or generated assessments.

Scheduling

Do not create recurring calls unless the learner explicitly requests them.

Before creating a recurring schedule:

  1. Confirm the schedule.
  2. Confirm the destination phone number.
  3. Explain that calls will be placed automatically.
  4. Confirm the learning topic.
  5. Provide a cancellation method.

Do not create duplicate schedules for the same learner, topic, and time window.

Cancellation

The learner must be able to cancel a scheduled recall call.

Cancellation should:

  • prevent future calls
  • preserve completed recall assessments
  • not delete previously collected learning results unless explicitly requested
Dry Run

Implementations should provide a dry-run or preview mode whenever possible.

In dry-run mode:

  • no real phone call is placed
  • the proposed call time is displayed
  • the topic is displayed
  • the proposed question flow is displayed
  • the assessment logic can be tested without contacting the learner

Dry-run mode should make it obvious that no real call will occur.

External Side Effects

An outbound phone call is an external side effect.

The implementation must make the call action visible to the learner and should provide enough information to understand:

  • when the call will occur
  • which phone number will be contacted
  • which topic will be assessed
  • whether the call is one-time or recurring

Do not silently place calls or silently create recurring schedules.

Sensitive Topics

This skill is intended for educational recall.

It must not present itself as a substitute for:

  • medical advice
  • legal advice
  • financial advice
  • emergency services

If a learner introduces a sensitive or emergency situation, the call should not attempt to provide professional or emergency intervention.

If the learner appears to require emergency assistance, encourage them to contact the appropriate local emergency service or qualified professional rather than attempting to handle the situation through this educational skill.

Error Handling

If required learning information is missing:

  • do not invent study material
  • explain what information is missing
  • request the required input through the host application

If the learner cannot answer a question:

  • do not shame them
  • allow them to continue
  • record the concept as a possible weak area when appropriate

If the phone call cannot be completed:

  • do not falsely report that the call occurred
  • return the appropriate failure status through the host application
  • preserve any completed assessment data

Example

Input:

text
topic: Hashing vs Encryption

study_context:
Hashing is generally a one-way transformation used to produce a fixed-length
digest. Encryption is designed to allow data to be recovered using the
appropriate key.

previous_score: 70

weak_areas:
- Hashing vs encryption

review_number: 2

Learner response:

"Hashing encrypts the password so that we can decrypt it later."

Assessment:

json
{
  "topic": "Hashing vs Encryption",
  "recall_score": 55,
  "status": "misconception_detected",
  "concepts_remembered": [
    "Hashing can be used in password-related systems"
  ],
  "weak_areas": [
    "Hashing vs encryption"
  ],
  "misconceptions": [
    {
      "concept": "Hashing vs encryption",
      "student_belief": "Hashing is reversible encryption.",
      "correct_understanding": "Hashing is generally one-way, while encryption is designed to allow data to be recovered using the appropriate key.",
      "evidence_from_response": "The learner described hashing as something that can be decrypted later."
    }
  ],
  "confidence": "medium",
  "summary": "The learner understands that hashing is relevant to password protection but is confusing hashing with reversible encryption.",
  "recommended_next_review": "1 day"
}

Implementation Notes

This skill defines the learning and conversation behavior.

The host application is responsible for:

  • obtaining user consent
  • securely handling phone numbers
  • securely handling provider credentials
  • scheduling calls
  • placing calls
  • cancellation
  • storing assessment results
  • preventing duplicate scheduled calls
  • implementing dry-run behavior
  • connecting the skill to a phone-call provider

The skill itself must not assume a specific phone-call provider.

Provider-specific authentication and call execution should remain in the host application.

© CALLE-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 5 other files (scripts, references) in skills/learning-recall-call of CALLE-AI/awesome-phone-call-agents.

  • SKILL.md
  • references/examples.md
  • references/safety.md
  • scripts/evaluator.py
  • scripts/llm_evaluator.py
  • scripts/test_recall.py

Open the folder on GitHubat commit 38d4118

Compare with similar skills

Learning Recall Call next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Learning Recall Call compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Learning Recall Call this skillCALLE-AI/awesome-phone-call-agents107—~3.2kAutomated safety check: PassMIT
Hermes Agent Skill AuthoringNousResearch/hermes-agent252k—~3.6kAutomated safety check: PassMIT
Configuring Oauth2 Authorization Flowmukul975/Anthropic-Cybersecurity-Skills34k—~1.7kAutomated safety check: PassApache-2.0
Authoring Skillsvercel/next.js143k—~1kAutomated safety check: PassMIT
PhoneBlockRunAI/ClawRouter6.6k—~2.4kAutomated safety check: PassMIT
Containing Active Breachmukul975/Anthropic-Cybersecurity-Skills34k—~2.7kAutomated safety check: PassApache-2.0

Similar skills

  • Hermes Agent Skill Authoring

    NousResearch/hermes-agent

    Author in-repo SKILL.md files: frontmatter and structure. An agent skill from NousResearch/hermes-agent.

    252k GitHub stars~3.6k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Configuring Oauth2 Authorization Flow

    mukul975/Anthropic-Cybersecurity-Skills

    Configures secure OAuth 2.0 authorization flows, including Authorization Code with PKCE, Client Credentials, and Device Authorization Grant, covering flow selection, PKCE implementation, token…

    34k GitHub stars~1.7k tokensUpdated 1 mo ago
    Backend & APIsAuto-check passed
  • Authoring Skills

    vercel/next.js

    Official

    How to create and maintain agent skills in .agents/skills/. An agent skill from vercel/next.js.

    143k GitHub stars~1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Phone

    BlockRunAI/ClawRouter

    Verify phone numbers (carrier + SIM-swap fraud signals) and place AI-powered outbound voice calls via BlockRun's gateway (Twilio + Bland.ai).

    6.6k GitHub stars~2.4k tokensUpdated 5 days ago
    Auto-check passed
  • Containing Active Breach

    mukul975/Anthropic-Cybersecurity-Skills

    Executes containment strategies to stop active adversary operations and prevent lateral movement during a confirmed security breach.

    34k GitHub stars~2.7k tokensUpdated 1 mo ago
    SecurityAuto-check passed
  • Abp Authorization

    abpframework/abp

    ABP permission system - PermissionDefinitionProvider, [Authorize] attribute, CheckPolicyAsync, IsGrantedAsync, ICurrentUser, IPermissionManager, multi-tenancy side.

    14k GitHub stars~1.3k tokensUpdated yesterday
    Backend & APIsAuto-check passed

More from CALLE-AI/awesome-phone-call-agents

All 101 skills in this repo
  • Accessible Outing Verifier

    CALLE-AI/awesome-phone-call-agents

    Demonstrates advisory accessibility-planning checks with offline fixtures and a proposed bounded CALL-E workflow; use for exploring unknown or qualified venue claims without making calls.

    107 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed
  • Ground Truth Gate

    CALLE-AI/awesome-phone-call-agents

    A skill your agent uses when an agent holds some evidence for a physical-world claim but the evidence is broader, narrower, or older than the exact question asked, and it must first decide whether a…

    107 GitHub stars~3.3k tokensUpdated yesterday
    Auto-check passed
  • Is It Accessible

    CALLE-AI/awesome-phone-call-agents

    Call a venue and ask the accessibility questions that matter to one specific person — step-free entry, hearing loop, guide dogs, quiet hours, changing places — then return a per-need verdict backed…

    107 GitHub stars~4.3k tokensUpdated yesterday
    Auto-check passed
  • Landmark Navigation Assist

    CALLE-AI/awesome-phone-call-agents

    Turns a pre-written, building-level location config into a CALL-E outbound phone-call task that guides a delivery driver through the last few hundred metres to a specific building using landmarks…

    107 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed
  • Research Gap Call Verifier

    CALLE-AI/awesome-phone-call-agents

    Turn cited business research into a bounded, approval-gated phone-call plan that asks only unresolved factual questions, then reconcile CALL-E-compatible results without treating voicemail, refusal…

    107 GitHub stars~1.7k tokensUpdated yesterday
    Auto-check passed
  • Structured Outcome Followup Call

    CALLE-AI/awesome-phone-call-agents

    Place a goal-driven CALL-E call that collects specific structured answers, score those answers against a deterministic rubric you supply, and conditionally trigger a follow-up action — all runnable…

    107 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed

Questions about Learning Recall Call

What does Learning Recall Call do?

Conducts an authorized phone-based active recall session with a learner, asks adaptive questions about a previously studied topic, identifies knowledge gaps and misconceptions, and returns a…. Learning Recall Call is an agent skill from CALLE-AI/awesome-phone-call-agents. Conducts an authorized phone-based active recall session with a learner, asks adaptive questions about a previously studied topic, identifies knowledge gaps and misconceptions, and returns a structured learning assessment.

How do I install Learning Recall Call in Claude Code?

Run `npx skills add CALLE-AI/awesome-phone-call-agents --skill learning-recall-call -a claude-code`. Or copy the skill folder (skills/learning-recall-call in CALLE-AI/awesome-phone-call-agents) into .claude/skills/learning-recall-call in your project. Claude Code loads it when a task matches its description.

How do I install Learning Recall Call in Codex?

Run `npx skills add CALLE-AI/awesome-phone-call-agents --skill learning-recall-call -a codex`. Or copy the skill folder (skills/learning-recall-call in CALLE-AI/awesome-phone-call-agents) into .agents/skills/learning-recall-call in your project. Codex loads it when a task matches its description.

Can I use Learning Recall Call in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add CALLE-AI/awesome-phone-call-agents --skill learning-recall-call -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/learning-recall-call, .gemini/skills/learning-recall-call, .github/skills/learning-recall-call and .opencode/skills/learning-recall-call in your project.

What does Learning Recall Call need to run?

Going by SKILL.md and its folder, Learning Recall Call needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Learning Recall Call access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Learning Recall Call safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Learning Recall Call use?

Learning Recall Call is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Learning Recall Call use?

About 3.2k tokens (SKILL.md is roughly 13k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2k tokens, read only when the agent opens those files.

What are the alternatives to Learning Recall Call?

Skills that share tags, products or a category with Learning Recall Call: Hermes Agent Skill Authoring (NousResearch/hermes-agent, 252k stars), Configuring Oauth2 Authorization Flow (mukul975/Anthropic-Cybersecurity-Skills, 34k stars), Authoring Skills (vercel/next.js, 143k stars) and Phone (BlockRunAI/ClawRouter, 6.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Learning Recall Call?

CALLE-AI (a GitHub organization) maintains it in CALLE-AI/awesome-phone-call-agents, which has 107 GitHub stars. The repository holds 101 skills in this directory. The repository was last updated on October 10, 2026.

Source: CALLE-AI/awesome-phone-call-agents on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.